Scott Shane Technology Strategy For Managers and Entrepreneurs Pearson

 

Technology Strategy for Managers and Entrepreneurs Scott A. Shane First Edition

 

Pearson New International Edition

Technology Strategy for Managers and Entrepreneurs Scott A. Shane First Edition

 

 

Pearson Education Limited Edinburgh Gate Harlow Essex CM20 2JE England and Associated Companies throughout the world Visit us on the World Wide Web at:  www.pearsoned.co.uk

© Pearson Education Limited 2014

All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without either the prior written permission of the publisher or a licence li cence permitting restricted copying in the United Kingdom issued by the Copyright Licensing Agency Ltd, Saffron House, 6–10 Kirby Street, London EC1N 8TS. All trademarks used herein are the property of their respective owners. The use of any trademark in this text does not vest in the author or publisher any trademark ownership rights in such trademarks, nor does the use of such trademarks imply any affiliation with or endorsement of this book by such owners. 

ISBN 10: 1-292-040321-292-04032-7 7 ISBN 10: 1 269 37450 8 ISBN 13: 978-1-292-04 978-1-292-04032-5 032-5 ISBN 13: 978 1 269 37450 7

British Library Cataloguing-in-Publication Data A catalogue record for this book is available from the British Library

 Printed in the United States of America

 

P

E

A

R

S

O

N

C

U

S

T

O

M

L

I

B

R

A

R

Y

 Ta  T able of Contents

1. Introduction Scott Shane

1

2. Technology Evolution Scott Shane

15

3. Technology Adoption and Diffusion Scott Shane

45

4. Sources of Innovation Scott Shane

71

5. Selecting Innovation Projects Scott Shane

97

6. Customer Needs Scott Shane

119

7. Product Development Scott Shane

147

8. Patents Scott Shane

175

9. Trade Secrets, Trademarks, and Copyrights Scott Shane

203

10. Capturing Value from Innovation Scott Shane

235

High-Tech ech Industries 11. Competitive Advantage in High-T Scott Shane

259

12. Technical Standards Scott Shane

281

13. Collaboration Strategies Scott Shane

301

I

 

Networked Industries 14. Strategy in Networked Scott Shane

329

15. Strategic Human Resource Management of Technical Professionals

II

Scott Shane

351

Index

375

 

 Introduc  Intr oduction tion Learning Objectives Introduction Technology Strategy: A Vig Vignette nette What Is Technological Innovation? Defining Technology Defining Innovation Defining Technological Innovation Why Technological Technological Innovation Is Important Getting Down to Business: Remembering Process Development

Why Technology Strategy? Purpose of the Text  Approach to Technology Technology Strategy Technology Strategy in Start-ups and Large, Established Firms Discussion Questions Key Terms Notes

Learning Objectives After reading this chapter, you should be able to: 1. Define technological innovation, explain why it is important, and

describe how firms use it to achieve their objectives. 2. Describe how technological innovation occurs, and explain the

effects that it has on individuals, firms, and society society.. 3. Define technology strategy , and explain why it is important for

entrepreneurs and managers to develop technology strategies. 4. Describe the approach to technology strategy taken in the text,

and explain why that approach is important. 5. Identify the core areas of technology strategy strategy,, and spell out how

these different areas of technology strategy affect businesses. 6. Describe how the core areas of technology strategy differ for new

and established firms, and explain why these types of firms need to take different approaches to technology strategy strategy..

From Chapter 1 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 1

 

Introduction

Technology Strategy: A Vignette1  The Xbox 360 game machine, developed by Microsoft to challenge Sony’s position as the market leader in the sale of video game consoles, is a prime example of a company’s effort to develop an innovative new product as part of its overall technology strategy. It also illustrates many of the topics that are covered in this text. Video game consoles are a high-technology product, requiring significant investment in research and development to create. Moreover, Microsoft’s effort to develop a competitive video game console involves a variety of strategic issues. First, video games involve network effects, which influence the relationship between game and console makers. If Microsoft can sell more Xbox 360 consoles than Sony can sell PlayStation 3 consoles, then video video game makers will design design the best games for the Xbox. The availability of the best games for the Xbox will increase sales of the consoles, generating a positive returns cycle for the company. Second, the manufacture of the consoles faces very significant economies of scale. As the volume of production of video game consoles goes up, costs per unit (the amount of money that the company needs to spend to create each console) decline. As a result, the largest producers of consoles have significant cost advantages over their competitors. Moreover,, initial sales are often made at a loss, as a company ramps up to Moreover minimum efficient scale. During that ramp-up process, companies often lose a lot of money. Microsoft, for example, lost $4 billion on the development and production of the Xbox hard drive and microprocessor alone.2 Third, success in this business depends on Microsoft’s ability to develop the right capabilities to manage a video game business. Traditionally, the company’s expertise has been in making computer software. However, video game consoles are pieces of computer hardware. Because the manufacture and sale of computer hardware are very different than the production and sale of computer software, to produce and successfully market the Xbox, Microsoft has to develop very different capabilities than it has needed in its software business. 3 Fourth, the production of the Xbox requires Microsoft to develop a new supply chain. The Xbox 360 contains 1,700 different parts, which all have to be brought together to make the product. To keep costs down, Microsoft produces the Xbox in China. It uses two factories, each of which serves as a hedge against problems that would shut down the other. Around these two core manufacturers are a host of suppliers of plastic parts for the boxes, capacitors, cooling fans, and other parts whose production needs to be coordinated.4 Fifth, the development and production of the Xbox requires Microsoft to manage contractual relationships with other companies. Specifically, it has to coordinate their production of the graphics chip and hard drive used in the Xbox. Its entire effort to sell Xboxes could derail if these producers run into production delays. While diversification protects Microsoft against problems with the production of the hard drive—the console’s hard drive is a commodity sourced from a variety of manufacturers—it is vulnerable to problems with the chip, which is custom-made by IBM. The reliance on IBM could prove to be a problem if anything hinders its ability to deliver the key component. 5

2

 

Introduction

I NTRODUCTION Technological innovation has become an important part of the process by which companies in many industries generate competitive advantage, making it a crucial part of firm strategy. strategy. In recent years, many companies have increased their level of  technological innovation to produce a greater variety of new products, and to introduce those new products to market faster. In many industries, the share of  sales and profits accounted for by products introduced in the past five years has been growing rapidly. rapidly. In fact, some companies, like 3M, now generate 40 percent of their sales from products that did not exist five years ago. Companies have also increased their level of technological innovation in response to competition. The reduction of costs and the improvement in quality of products made in lower wage countries, like China and India, have posed a major challenge for firms in developed countries, like the United States and Germany. Many firms from developed countries have responded to this challenge by introducing new products at a faster pace to stay ahead of imitators, and by using technological innovation to reduce their own production costs.6 Technological innovation has also increased as more companies that once produced commodity products now seek to differentiate their offerings from those of  competitors. The desire of more companies to offer differentiated products hasinshortened the product life cycle and has increased the importance of investments new 7 product and process development. Furthermore, technological innovation has increased as companies have turned intellectual property into a marketable asset. In recent years, the licensing of technology to other companies has become an important revenue stream for many companies, with some, like IBM, adopting the approach that all of its intellectual property is potentially for sale. This marks a major change from only a couple of  decades ago when intellectual property was used only as an input into a company’s product or service. In addition, there has been significant growth recently in the formation of hightechnology start-ups that use funding by venture capitalists and business angels (individuals who invest their own money in start-up companies, usually by taking an equity stake in them) to introduce high-technology products and compete with established firms. As a result, technological technological innovation has also been increasing  because of entrepreneurs, entrepreneurs, including those who create spin-off companies, using technology developed at major corporations and universities. This emphasis on technological innovation as a way to generate or preserve competitive advantage has led to an increased need for managers and entrepreneurs who can develop strategies to successfully manage this activity. While companies can, and do, introduce new products, improve production processes, and target new markets without strategies or plans, companies are better at these activities if they develop, and execute, an effective strategy to undertake them. 8 By combining an understanding of markets and technological evolution with an understanding of firm organization and capabilities in a deliberate and organized manner, managers and entrepreneurs can generate value by developing technology products and services that better meet customer needs, and can become better at capturing that value. The increased need for managers and entrepreneurs to develop strategies for technological innovation, in turn, has led to an increased demand for business school courses in the management of technological innovation, and for strategic

3

 

Introduction

management courses that focus on issues particular to high-tech companies. In short, technology strategy is now an important part of the education of business leaders.

W HAT I S T ECHNOLOGICAL I NNOVATION? The previous vignette illustrates the importance of understanding technology strategy by highlighting many of the important issues that companies face and that are the subject matter of this text. But before we get into a discussion of what technology strategy is, and how to develop it effectively, effectively, we need to lay a little ground work. We have to define technological innovation and explain how firms use it to achieve their objectives. After all, strategy is just an approach to achieving a particular goal, making technology strategy nothing more than an approach to using technological innovation to achieve a goal. So our first step is to define technological innovation. innovation. That is best done  by dec decom ompos pos in ing g the ph phras ras e int o its two par ts, the co conce ncept ptss of technology and innovation.. innovation

Defining Technology While there has been a tendency for the popular media to use the word technology as shorthand for information technology, technology, technology is much broader than just information technology. It is the application of tools, materials, processes, and techniques to human activity. Certainly, information technology—the use of zeros and ones in digital form on computers—is an important technology, but there are many other important technologies as well. Biologically based technologies, such as those used to create new drugs or to clean up pollution, are also important. Similarly, mechanically based technologies, such as those that make pumps or valves, matter. New materials, such as those in new ceramic composites, are valuable too. So when this text discusses technology, we aren’t talking just about information technology. Rather, we are talking about a host of technologies, including, but not limited to, new microorganisms, new mechanical devices, new materials, and a variety of other products and processes. So when you see the word technology or the innovation, do not just think of the Internet and computer softphrase technological innovation, ware, think of processes like nanofabrication (the process of making things less than one micrometer in size) and products like fuel cells, ceramic composites, new drugs, or heart valves. The use of technology is more prevalent in some industries than in others. Figure 1 shows the industries that the U.S. government defines as technologyintensive. Clearly, these industries are the ones in which an understanding of  technology strategy is important to entrepreneurs and managers.

4

 

Introduction

FIGURE 1

Technology-Intensive Industries

Industry Aerospace product and parts manufacturing Agriculture, construction, and and mining machinery manufacturing All other electrical equipment component manufacturing Architectural, engineering, and related services Audio and video equipment manufacturing Basic chemical manufacturing Commercial and service industry machinery manufacturing Communications equipment manufacturing Computer and office machine repair and maintenance Computer and peripheral equipment manufacturing Computer systems design and related services Data processing services Educational support services Electrical equipment manufacturing Engine, turbine, and power transmission equipment manufacturing Industrial machinery manufacturing Management, scientific, and technical consulting services Manufacturing and reproducing and optical media Medical equipment and supplies magnetic manufacturing Motor vehicle body and trailer manufacturing Motor vehicle manufacturing Motor vehicle parts manufacturing Navigational, measuring, electromedical, and control instruments manufacturing Online information services Ordnance & accessories manufacturing—ammunition (except small arms) manufacturing Ordnance & accessories manufacturing—other ordnance and accessories manufacturing Ordnance & accessories manufacturing—small arms ammunition manufacturing Ordnance & accessories manufacturing—small arms manufacturing Other chemical product and preparation manufacturing Other general purpose machinery manufacturing Paint, coating, and adhesive manufacturing Pesticide, fertilizer, and other agricultural chemical manufacturing Petroleum refineries Pharmaceutical medicine manufacturing Resin, synthetic and rubber, and artificial and synthetic fibers and filaments manufacturing Scientific research and development services Semiconductor and other electronic component manufacturing Soap, cleaning compound, and toilet preparation manufacturing Software publishers

The U.S. government defines these industries as technology intensive because the firms in them devote a large proportion of  their revenue to research and development. http://www.nsf.gov/statistics/seind06. Source: Adapted from Science and Engineering Indicators, 2006, http://www.nsf.gov/statistics/seind06.

Defining Innovation The next important definition is that of innovation—the process of using knowledge to solve a problem. Innovation is different from invention, which is the discovery of  a new idea, because it involves more than just coming up with an idea about how to

5

 

Introduction

use knowledge to solve a problem. For example, during the Renaissance, inventors came up with the ideas for parachutes, fountain pens, mechanical calculators, and  ball bearin bearings. gs. However, these ideas did not become innova innovations tions until much later  because they were not technically feasible and could not be implemented impleme nted at the time 9 that the ideas were discovered.

Defining Technological Innovation So what is technological innovation? Simply put, it is the use of knowledge to apply tools, materials, processes, and techniques to come up with new solutions to problems.10 Some innovations, like Michael Dell’s approach to selling computers—selling personal computers assembled from standard components direct to customers11— were not technological innovations because the knowledge used to solve a problem did not involve new technology (tools, materials, processes, and techniques), but, instead, involved new ways of organizing a business. However, other innovations, like genetic engineering, were technological innovations because the knowledge used to solve problems involved new tools, materials, processes, and techniques. Technological innovation can be planned or accidental. Sometimes the use of  technical knowledge to solve problems is purposeful. For instance, many companies invest in research and development with the goal of coming up with an innovative new product or process that will give them an advantage over their competitors. However, other times technological innovation isn’t the result of a deliberate attempt to solve a particular problem. For example, example , Pfizer was not looking for a solution to the problem of erectile dysfunction when it came up with Viagra. The solution to this problem was merely a side effect discovered in tests of the drug for its intended purpose of treating angina in cardiac patients. Similarly, Alexander Fleming discovered penicillin because a spore of mold contaminated a sample of   bacteria that he was using, and inhibited its growth. 12 This definition of technological innovation has three important implications for understanding underst anding technology strategy. strategy. First, as Figure 2 shows, technological technological innovation doesn’t have to be profitable. Companies can, and do, come up with solutions to problems that they can’t make any money exploiting. So a big part of technology strategy is coming up with ways to make money from technological innovation (and another big part of technology strategy is keeping that money, rather than letting competitors take it). FIGURE 2 What Is Technological Innovation?

A technological innovation involves the application application of knowledge to solve a problem, and need not work nor result in something of commercial value. Conley, March 17, 2007; Distributed by UFS, Inc. Source: Darby Conley,

6

 

Introduction

Second, there are many different kinds of technological innovations. They can come from any kind of technical knowledge, from knowledge of computer science to knowledge of biology to knowledge of new materials. Moreover, technological innovation can take the form of solutions to a variety of different kinds of problems, from speeding production to introducing new products that meet customer needs to facilitating distri bution.  butio n. The developmen developmentt of a new materi material al for making making aircraft aircraft lighter lighter or or a new way way to produce biodiesel are just as much technological innovations i nnovations as Windows Vista. Third, there is not a direct, one-to-one, relationship between technological change and new products or processes. Some new technologies may lead to only one, or even no, new products or processes, while others might make possible a very large number of them. Take, for example, the case of radio frequency identification (RFID) technology. This technology has led to the creation of ExxonMobil’s speedpass, the E-ZPass highway toll collection system, a system for inventory tracking in libraries, a way to identify the parts and components in computer and automobile manufacturing, a method to record quality problems in disk drive manufacturing, and a way to track products p roducts shipped from manufacturers to retailers. 13

Why Technological Technological Innovation Is Important Technological innovation is an important source of value creation. The application of  knowledge to human activity allows for the more efficient production of existing products and services (that is, production that takes less money or effort) and allows for the creation of products and services that meet needs that were not previously satisfied.14 For example, technological innovation allows us to provide burn victims with artificial skin, and makes the cost of heating our homes cheaper. Because of the importance of technological innovation to value creation, technological change has tremendous economic impact. Economists have shown that much of the growth in gross domestic product comes from the use of technology to make more productive use of labor and capital.15 Technological innovation is also important because it has a profound effect on the creation of wealth for individual entrepreneurs and corporate shareholders. New technology makes possible the formation and rapid growth of companies like Google and Microsoft that were not around four decades ago, and attributes to the death of  leading companies of a previous era, such as Digital Equipment Corporation. In fact, researchers have shown that much of the wealth creation by entrepreneurs is accounted for by the creation of products products and services based on technological 16 innovation. Technological innovation is also important because it has a major impact on our lives. For instance, the way in which we look for information has been dramatically altered by Internet search, how we get around was forever changed by the airplane and the automobile, and life expectancy has been greatly lengthened by medical diagnostics and treatments. Not all of the impact of technological innovation is good. Research has shown that the rapid and constant technological progression that we have experienced in recent years causes social isolation, increased time pressure, the constant need to multitask, the inability to disconnect from work, and other adverse consequences. As Figur Figuree 3 illustr illustrates, ates, somewhat somewhat humorously humorously,, many people have found the loss of  privacy and control that technological innovation has generated sometimes outweighs the benefits of the opportunities that technological change has created. Designing an effective technology strategy entails the creation of new technology products and processes that provide benefit to your firm while minimizing the

7

 

Introduction

FIGURE 3 Technological Change Isn’t Always Good

Many new communications technologies have an adverse effect on personal privacy. privacy. Source: Berkeleybreathed.com, July 10, 2005, http://cartoonistgroup.com/store/add.php?iid=11027.

adverse effects of technological change on your customers and society at large. Many companies have run into trouble when they have made use of technology to achieve a competitive advantage and that technology has had adverse consequences. For example, the efforts of retailers T.J. Maxx and Marshalls to gain competitive advantage by using information technology to gather customer information that could be used for targeted marketing led to a loss of customer information to identity thieves who could perpetuate fraud against tens of millions of people, alienating large num bers of their customers.17

GETTING DOWN TO BUSINESS Remembering Process Development18 Sometimes, process development—the anticipation and resolution of problems that arise in the production of a product—is more important than product development as a source of competitive advantage for high-technology firms. The Sensor razor, introduced by Gillette in 1990, is one such example. This product was a new razor that provided a much better shave than alternative razors. The Sensor’s advantage in providing a close shave was its use of 23 floating parts, rather than the 5 floating parts of which most other razors are composed.

8

The idea behind the Sensor razor was not particularly surprising. Most engineers working on developing razors at the companies making them knew that the more floating parts you could put on the razor ra zor,, the closer a shave it would provide. Although the engineers at many companies knew this about razor design, no other companies developed a razor with more than 5 floating parts. Why? Increasing the number of floating parts on a razor generate generated d significant production production challenges. challenges. The

 

Introduction

mounting of so many parts on the razor bar could not damaged in the process of attaching them to the metal  be done done by the the standard standard approa approach ch of of gluing. gluing. To To develop develop  bar, they needed to weld them withou withoutt using heat. the Sensor and make it work effectively, Gillette’s engi- This, in turn, required them to create a non-heatneers had to figure out how to mold the plastic blades producing laser spot welder that could make very fast into the cartridge. And, because the plastic that Gillette used for other razors lost bounce with repeated use, they had to shift to a new resin to make this work. In addition, the engineers needed to have the  blades float on separate springs to get the close shave that would make Sensor better than other razors. This was difficult to do because the blades were as thin as a piece of paper. To To make the blades float on the springs, the engineers had to attach the blades to a metal support bar. Then, to keep the blade edges from getting

intricate welds with very low tolerances. All of this process development made it possible for the company to develop an innovative product that met customer needs better than the alternatives offered  by competitors. Moreover Moreover,, Gillette’s process development provided the company with a better manufacturing process than its competitors, thus creating an additional competitive advantage. In short, Gillette’s process development capabilities drove the development of an important new product, the Sensor razor.

 Key Points

• Technological innovation is the use of technical knowledge to come up with solutions to problems. • Technological innovation is important to entrepreneurs and managers because it provides a mechanism to create and preserve competitive advantage.

W HY T ECHNOLOGY S TRATEGY? While the previous section of this chapter discussed the importance of technological innovation to companies and outlined some basic forms that technological innovation can take, it did not say anything about why companies need to develop a strategy to manage innovation. However, many observers believe that companies can, and should, develop technology strategies. A technology strategy is the approach that a firm takes to obtaining and using technology to achieve a new competitive advantage, or to defend an existing technology-oriented competitive advantage against erosion. Technology strategy is different from overall business strategy in several important ways: 1. Technology strategy has to deal much more with issues is sues of uncertainty than general business strategy because technological change is highly uncertain. A very small portion of new technology ideas result in new products or processes. 2. Technology strategy involves the use of intellectual property management to capture financial returns to a much greater degree than general business strategy. 3. Technology strategy involves the creation of new products and services s ervices that are sometimes new to the world, which demand different mechanisms for assessing market needs and designing products than is the case with general business strategy. 4. Technological change occurs in ways that influence the design desi gn of effective strategies, and create business dynamics that are different from those that exist when new technology is not important.

9

 

Introduction

5. Organizations faced with a great deal of technical change need to be structur structured, ed, manage human resources, and design business models in ways that are different from organizations that are not faced with technical change. 6. Making decisions about technology projects requires the use of different decision-making tools than is the case with nontechnology projects. 7. Technological change opens up opportunities for new, high growth businesses in ways not possible in other settings. 8. High-technology businesses face many strategic issues, like standards and increasing returns, that are much rarer with low technology businesses.

In short, technology strategy differs from general business strategies in enough ways to make a text (and a course) focused on it useful to students.

Purpose of the Text This text will provide you with analytical tools that will help you to become a successful technology strategist. Using these tools will require you to understand a variety of things, including the sources of opportunity for technological innovation and how to identify them; the assessment of customer needs; the product development process; technology evolution; new product adoption and diffusion; increasing returns, network effects, technical standards; standard s; tradeprotection; secrets, trademarks, copyrights, patents, and other and types of intellectual property firm capabilities and competitive advantage; management of the innovation process; and organizational form and structure.

Approach to Technology Strategy Four observations about the approach to technology strategy taken in this text are important to make. First, this text is inherently interdisciplinary. Technological innovation is a general management problem whose solution requires an understanding of psychology, economics, strategy, finance, organization behavior, and marketing. Therefore, the concepts that are discussed are drawn from a wide variety of fields, including economics, psychology, sociology, strategic management, finance, and operations management. Second, this text discusses technology strategy in all types of businesses in all kinds of industries, from aerospace to environmental consulting to motor vehicles. This means that the issues that are discussed, and the examples that are given, are not limited to those industries commonly thought of as “high tech,” like computer software. As a result, the focus is on issues that transcend industries, like the management of intellectual property and the use of incentives to motivate people to come up with and implement new ideas. Third, the text discusses technology strategy in service businesses as well as manufacturing businesses. Technology strategy for services is different from technology strategy for products because services are intangible and cannot be examined  before custom customers ers purchas purchasee them and are insep inseparable arable—they —they are used in the same 19 place that they are produced. As a result, innovation in services involves greater  joint activity between companies and their customers, has goals that are more difficult to measure, is more difficult to standardize, and involves outputs that cannot be maintained in inventory inventory..20

10

 

Introduction

Coverage of service businesses is important because technological innovation is not something that only manufacturing manufacturing firms undertake. undertake. In fact, research and development (R&D) spending in services now accounts for 30 percent of total R&D expenditures. Moreover, services are a growing part of the economies of most developed countries. In the United States, for instance, services now account for 68 percent of the gross domestic product. Furthermore, in some industries, like biotechnology and software, software, technological innovation innovation in services is central central to competitive advantage. Fourth, this text examines technology strategy in the context in which it occurs. Technological innovation does not occur in a vacuum, but, instead, is affected by the environment in which it occurs. Therefore, it considers the effects of the industrial, geographic, political, regulatory, cultural, competitive, and economic context when discussing different aspects of o f technology strategy.

Technology Strategy in Start-ups and Large, La rge, Established Firms While technological innovations are often developed and exploited by large, established corporations, sometimes people start new companies to create and exploit these innovations, making the creation of new companies a very important mechanism through which technological get exploited. Forbyexample, 2003, researchers estimated that just thoseinnovations start-up companies funded venture in capitalists since 1970 employed 10 million people, or 9.4 percent of the private sector labor force in the United States, and generated $1.8 trillion in sales, or 9.6 percent of   business sales in this country.21 In 2000, the 2,180 publicly owned companies that received venture-capital backing between 1972 and 2000 comprised 20 percent of all public companies in the United States, 11 percent of their sales, 13 percent of their profits, 6 percent of their employees, and one-third of their market value, a figure in excess of $2.7 trillion dollars.22 However, most textbooks approach the topic of technology strategy exclusively from the perspective of the large, established firm. This perspective is limited for those students who want to work in young companies or even start one of those companies some day. It is also limited for those students who want to work for large companies because those companies often need to compete with, or partner with, small, start-up companies. This discussion takes a different perspective. It looks at technology strategy issues from the perspective of both the small, new firm and the large, established one. For many aspects of technology strategy, this only means clarifying that the strategy is the same for all firms, whether they are large and established or small and new. For example, the demand for technical standards to permit different companies to link their components together is true regardless of whether an industry is composed primarily of large, established firms, or small, new ones. However, for other aspects of technology strategy, this means discussing two approaches to the same issue. For instance, the way in which large, established companies with deep pockets protect their intellectual property is not the same as that used by cash-poor small, young businesses. Where aspects of technology strategy are different for large, established firms and small, new companies, the text specifically identifies and discusses those differences. For example, as Figure 4 shows, R&D intensity (the amount of money spent on

11

 

Introduction

FIGURE 4 R&D Intensity by Firm Size for the Period 1999–2003

12.00   y    t    i   s   n   e    t   n    I     D    &    R

10.00 8.00 6.00 4.00 2.00 0.00

  2  4    5  –

  9   –  4    5   2

  9   9   9   9   9   9   e   9  9   o  r   –  9   2  4   4  9   9  9   9  9   9  9   9  ,   ,   ,    0   –   –   –   m   4   9   4      5   0   0   –   –   2    5  0   o  r   1  0   2    5  0   0  0   0  0   0  –   0    0   0   0   0  ,   ,    0   0   1    5  ,     5 ,    1  0   2 Firm Size

Technology strategy is different in large and small firms for many reasons, including the tendency of small firms to spend more of their revenue on R&D than large firms. Source: Compiled from data in Science and Engineering Indicators, 2006, http://www.nsf.gov/statistics/seind06/c4/c4s3.htm.

R&D as a percentage of sales) is much higher for small firms than for large ones. This means that the approach to generating competitive advantage in new, small firms depends much more on the use of legal barriers to imitation and products with attributes that competitors cannot deliver than is the case for established, large firms.

 Key Points

• Technology strategy is the approach that firms take to obtaining and using technology to achieve a new competitive advantage or to defend an existing competitive advantage against erosion. • Technology strategy helps companies in many industries i ndustries to perform better bet ter.. • Technology strategy involves a variety of issues, which can be categorized as those related to understanding technical change, meeting the needs of customers, capturing the value created from innovation, and implementing a technology strategy.

DISCUSSION QUESTIONS

12

1. What is technology? What are some examples of dif-

4. What are some key aspects of technology strategy

ferent types of new technology? What do they have in common? How are they different? 2. What is technological innovation? Why is it important to firms? 3. What is technology strategy? Why do firms in hightechnology industries need to develop technology strategies?

that entrepreneurs and managers managers need to learn? Why are these aspects central to technology strategy?

 

Introduction

KEY TERMS Appropriate: To capture the returns from investment in innovation and keep them from going to competitors. Cost per Unit: The amount of money that a company

Technological Innovation: The use application of  tools, materials, processes, and techniques to come up with new solutions to problems.

needs to spend to create one of a product. Innovation: The process of using knowledge to solve a problem. Invention: The discovery of a new technical idea. Process Development: The anticipation and resolution of problems that arise in the production of a product.

Technology: The application of tools, materials, processes, and techniques. Technology Strategy: The approach of firms to obtaining and using technology to achieve a new competitive advantage or to defend an existing competitive advantage against imitation.

NOTES 1. Adapted from Guth, R. 2005. Getting Xbox Xbox 360 to market. Wall Street Journal, November 18: B1, B5. 2. Ibi Ibid. d. 3. Ibi Ibid. d. 4. Ibi Ibid. d.

14. Mokyr Mokyr,, J. 1990. The Lever of Riches . New York: Oxford University Press. 15. Schil Schilling, ling, M. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill Irwin.

5. Ibid. d.lling, M. 2005. Strategic Management of Technological 6. Ibi Schilling, Schi Innovation. New York: McGraw-Hill Irwin. 7. Allen, K. K. 2003. Bringing New Technology Technology to Market. Upper Saddle River, NJ: Prentice Hall. 8. Clark, K., and T. T. Fujimoto. 1991. Product Development Performance. Boston: Harvard Business School Press. 9. Mokyr Mokyr,, J. 1990. The Lever of Riches . New York: Oxford University Press. 10. Afuah Afuah,, A. 2003. Innovation Management. New York: Oxford University Press. 11. http://en.wikiped http://en.wikipedia.org/wiki/Dell#Origins_and_ ia.org/wiki/Dell#Origins_and_ evolution 12. Gilbert, S. 2006. The accidental innovator. innovator. Working Knowledge, July 5, http://hbswk.hbs.edu/item/ http://hbswk.hbs.edu/item/ 5441.html.

16. Eckhardt, J. 2003. the Weak Acquire Wealth: An Examination of the When Distribution of High Growth Startups in the U.S. Economy, Ph.D. Dissertation, University of Maryland. 17. Sidel, R. 2007. Giant retailer retailer reveals customer data  breach. Wall Street Journal, January 18: D1, D6. 18. Adapted from Hammonds, Hammonds, K. 1990. How a $4 razor ends up costing $300 million. Business Week , 3143: 62–63. 19. Mohr Mohr,, J., S. Sengupta, and S. Slater. Slater. 2005. Marketing of High Technology Products and Innovations (2nd edition), Upper Saddle River, NJ: Prentice Hall. 20. Ettli Ettlie, e, J. 2000. Managing Technological Technological Innovation . New York: John Wiley Wiley.. 21. Venture Impact 2004, http://www http://www.nvca.org/pdf/ .nvca.org/pdf/ VentureImpact2004.pdf.

13. Srivastava, B. 2004. Radio frequency technology: technology: The next revolution in SCM. BusinessID Horizons , 47(6):

22. Gompers, P., and P., J. Lerner. 2001. The Money of  . Invention: How Venture Capital Creates New Wealth

60–68.

Boston: Harvard Business School Publishing.

13

 

14

 

Technology Evolution

From Chapter 2 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 15

 

Technology Evolution Learning Objectives Technology Evolution: A Vignet Vignette te Introduction Evolutionary Patterns of Development Radical and Incremental Technological Change Technology S-Curves Shifting S-Curves Who Shifts the S-Curve? Using S-Curves as a Management Tool The Abernathy-Utterback Model The Nature of Innovation and Competition New Firm Performance Number of Firms in the Industry Limitations to the Model

Getting Down to Business: The History of  Electric Vehicles Modifications to the Abernathy-Utterback Model Reverse Product Cycle Theory Competence-Enhancing and CompetenceDestroying Innovation  Architectural Innovation Disruptive Technology and Value Networks Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Explain

why technology tends to develop in an evolutionary

manner. a technology S-curve and describe what it measures.

2. Graph

3. Define

a shift in the S-curve, and explain who shifts the S-curve.

4. Spell

out the pros and cons of using technology S-curves as a management tool.

5. Describe

the Abernathy-Utterback model of technology evolution and explain how it is useful to technology strategy. strategy.

6. Define

a dominant design and interpret its effect on technology development and industry competition.

16

 

Technology Evolution

7. Define

radical and incremental technological change and explain how the two types of innovation affect technology strategy.

8. Summarize

the modifications that researchers have made to the Abernathy-Utterback model, and explain why they were needed, and how they are helpful.

9. Describe

architectural innovation and explain how it influences technology strategy.

10. Identify

the types of innovation that new and established firms are each better suited to develop, and explain why this is the case.

Technology Evolution: A Vignette1  The history of photography provides a good example ex ample of technology evolution. In 1839, the first photographic technique, the daguerreotype, was developed in France. This technology produced images on copper plates that had been sensitized s ensitized and coated with silver. In the 1850s, a new techte chnology for photographic images was developed. With it, a photographer would use silver nitrate to condition a glass plate coated with a substance called collodion just before taking a photograph. By exposing the glass to light, a negative image would be developed and then fixed on photosensitive paper through exposure to sunlight. This technology, technology, called “wet plate,” was followed by the development of “dry plate” technology te chnology in the 1870s. With dry plate technology technology,, photosensitized glass plates were covered with a dry gelatin emulsion. Because dry plates could be produced in factories, they could be made on a larger scale and at a lower cost than wet plates. Therefore, dry plate technology was an important commercial advance. Celluloid roll film was the next technology to be developed. In 1889, George Eastman introduced a photosensitive celluloid ce lluloid film that was rolled into the back of a camera. The development of roll film led to a dramatic growth in the market because it allowed ordinary people to take photographs. On the back of this technology, technology, Kodak grew from a tiny start-up into a multinational corporation. Celluloid roll film technology remained the dominant technology in cameras for approximately 100 years. But in the 1990s, Apple and Sony both introduced cameras that used digital images that could be viewed without any processing or developing. This example describes the evolution of a single technology, technology, and points out some of the important aspects of technology evolution that will be discussed in this chapter—the incremental nature of most technological innovation and the tendency of radical technological changes to be introduced by firms outside the industry.

17

 

Technology Evolution

FIGURE 1

Technological Changes That Kodak Has Faced

While exaggerating to make the point poi nt humorously, humorously, the figure shows that Kodak has faced waves of  technological change. Source: www.cartoonstock.com.

I NTRODUCTION New technologies tend to evolve in ways that allow you to use tools and techniques to forecast their evolution. By forecasting technology evolution, you can develop a strategy that considers the emergence of technologies, changes in the dominant strategies and structures of other firms, and shifts in the nature of competition between firms.2 This chapter provides a framework for understanding the evolution of technology, and presents several useful tools fordevelopment. managing it. The section describes the evolutionary pattern of technological Thefirst second section defines radical and incremental technological change. The third section discusses Foster’s technology S-curve. The fourth section describes the Abernathy-Utterback model of technology evolution. The fifth section discusses the modifications that different scholars have made to the basic Abernathy-Utterback model to incorporate the reverse product cycle, architectural innovation, disruptive innovation, and competence-enhancing and competence-destroying innovation.

E VOLUTIONARY  PATTERNS

OF   D EVELOPMENT

Many researchers view technological advance as a process that begins with basic scientific discoveries and ends with commercial products that are adopted by a wide range of customers. When new technology is first invented, it emerges in a primitive form in which its usefulness is unclear, often even to the inventors of the technology technology..

18

 

Technology Evolution

Over time, as researchers work on the technology, it evolves and develops, with its performance improving to the point at which someone identifies a commercial use for it, and introduces a product based on the technology to the market. The companies developing products based on the technology then refine it, and introduce additional generations of products that are based on improved versions of the technology. As the technology gets better, it meets the needs of a wider range of  potential customers, and products based on the technology are adopted by more and more users. This technological evolution depends on the process through which scientific advance occurs. New advances are made as researchers seek answers to current technical problems, building on prior knowledge that has accumulated. 3 This process leads scientists and engineers to work within particular frameworks, or paradigms, which influence how technical problem solving occurs. These frameworks affect both the choice of problems that scientists and engineers work on and the ways that they go about answering them. For example, the invention of streptomycin in the 1940s led to the development of a paradigm of developing pharmaceuticals through synthesis and subsequent testing of large numbers of organic molecules.4 That is, after this paradigm was developed, scientists in all pharmaceutical firms focused on synthesis and testing tes ting of organic molecules to develop new drugs untilThe a new paradigm emerged. is not entirely driven by scientific advance. Social, evolution of technology economic, and political forces affect the path that technological advance takes, each in somewhat different ways. The economic environment influences the nature of incentives created for technological advance, thus affecting the evolutionary path by increasing the rewards for development to go in certain directions and not in others. Political forces have an effect because they influence the rules of the society in which the technological advance takes place. As a result, they lead the technology to develop in ways consistent with society’s rules and not in ways inconsistent with it. Social forces matter because technological development takes place in a human environment in which the fears, motivations, and attitudes of people come into play. Consequently,, technology evolution tends to move toward things that people are supConsequently portive and accepting of and away from things that they are afraid of or intolerant of. 5 The tendency for scientists and engineers to work within technological paradigms leads to the creation of technology trajectories, which researchers define as paths of improvement of a technology on some performance dimension. 6 Take, for example, a key technological trajectory in the computer industry—the trend toward smaller and more powerful microprocessors. In 1965, Gordon Moore, the founder of  Intel, made a statement that has become known as Moore’s Law. He argued that the number of transistors that could be packed onto an integrated circuit at the minimum cost for the component will double every two years. As Figure 2 shows, over time, engineers have developed better ways of packing more processors onto each microchip, and the chips have become more powerful along the trajectory laid out in Moore’s law. Although technology paradigms have a valuable focusing effect on research, they also have an important downside. They tend to limit the alternatives that researchers are willing to consider when problem solving, and, consequently consequently,, often keep researchers from identifying fundamentally different, and better, alternatives.7 For example, early on in his career, Robert Langer, an MIT chemical engineering professor and one of America’s most prolific inventors, developed a new type of plastic that could be used to deliver a large molecule drug. When Langer presented this finding to the scientific advisory board of a leading pharmaceutical

19

 

Technology Evolution

FIGURE 2

The Technological Trajectory in Microprocessors

10,000,000,000

Number of Transistors Transistors Doubling Every 18 Months    t    i   u   c   r    i    C    d   e    t   a   r   g   e    t   n    I   n   a   n   o   s   r   o    t   s    i   s   n   a   r    T    f   o   r   e    b   m   u    N

Itanium 2 (9 MB cache)

1,000,000,000

Itanium 2 100,000,000

Number of Transistors Transistors Doubling Every 24 Months

Pentium 4

Itanium Pentium III Pentium II

10,000,000

Pentium

1,000,000

486 386

100,000

286 8086

10,000 8080 2,300

4004  8008 1971

1980

1990

2000

2004

Year

Moore’s Law is perhaps the most famous example of a technological trajectory; i t indicates that the number of transistors on an integrated circuit has doubled every two years since the 1970s. Source: http://en.wikipedia.org/wiki/Image:Moore_Law_diagram_%282004%29.png.

company, two Nobel-prize-winning members of that board responded that Langer’s discovery was impossible. Similarly, when Langer presented the same finding to a conference of polymer chemists, they responded that what he had just presented was contrary to the established literature, and was just plain wrong. However, Langer went on to garner many patents and found several companies  based on this discove discovery ry,, which subseque subsequently ntly was conside considered red a paradig paradigm-shifti m-shifting ng 8  breakthrough in drug delive delivery ry.. As the previous paragraph alludes, at discrete points in time, opportunities appear to fundamentally change the technological paradigms within which scientists and engineers operate. For example, the opening vignette in this chapter describes the paradigm shift to digital technology from chemically based film in the 1990s. The identification of these paradigm shifts is very important i mportant because they fundamentally alter the ways in which products and services are developed, and so have profound effects on technology strategy. Unfortunately, identifying paradigm shifts is much more difficult than Figure 3 suggests. Innovators often don’t know in advance if they have come up with paradigm shifting technologies or if the reason that the new technologies are poorly received is just that they are bad ideas. Take, for example, the Segway Human Transporter, a device that famed venture capitalist John Doerr said would be more

20

 

Technology Evolution

FIGURE 3

Paradigm Shifts

Technology entrepreneurs and managers wish that the identification of paradigm shifts were this easy; in reality, reality, they are very difficult to spot in advance. Source: New Yorker Magazine. 1975. The New York Album of Drawings, 1925–1975. 1925–1975. New York: Penguin Books.

important than the development of the Internet. While the founder of this device, Dean Kamen, thought it would shift the paradigm in how human beings get around, the Segway transporter transporter turned out to be inferior to foot power, electric vehicles, scooters, and mopeds for most of the users for which it was intended. As a result, despite the near $100 million invested in the technology, only a couple thousand of the devices were ever developed, and the device did not change the transportation paradigm.9

R ADICAL AND   I NCREMENTAL T ECHNOLOGICAL C HANGE Most technological innovation is incremental, and involves small improvements to existing technologies. A good example of an incremental incremental innovation is Intel’s 486 microprocessor. This microprocessor was faster than previous generations of Intel’s products, but the changes that made it faster represented relatively small improvements to its i ts basic technology. Other times, technological innovation is radical, and involves a fundamentally new way of solving a problem. A good example of a radical technology is the digital camera, which is based on fundamentally different technical principles from traditional chemical film fil m technology. While research has shown that only a small percentage of all new technologies are radical,10 they often have enormous impact (see Figure 4) because they are high risk–high return developments. Moreover, their creation and implementation requires fundamentally different strategies and organizational structures from those used for more incremental products. 11 For example, the techniques for gathering 12 that work market datainnovations and the tools forpoorly making about projects well for research incremental work fordecisions radical innovations.

21

 

Technology Evolution

FIGURE 4

Radical Technologies That Have Changed Markets

Aspirin Automobile Beta Blockers Computer

Internet Jet Airplane Medical Diagnostic Imaging

Personal Digital Assistant Superconductor Television

Operating Systems Digital Music Fiber Optics Genetic Engineering

Mobile Telephone Microprocessor Organic Fibers Personal Computer

Transistor Video Cassette Recorder Xerography

Some new technologies have dramatically changed markets because they were based on fundamentally different ways of solving problems. Sources: Adapted and compiled from Rothaermel, F. 2000. Technological Technological discontinuities and the nature of  competition. Technology Analysis and Strategic Management, Management, 12(2): 149–160; Brown, S., and Eisenhardt, K. 1997. The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative organizations. Administrative Science Science Quarterly, Quarterly, 42(1): 1–34; Christiansen, C. 1998. Valurec’s venture into metal injection injec tion molding. Harvard Business School Teaching Note, Number 698–002; Miller, W. W. 2006. Innovation rules! Research Technology Management, Management, 49(2): 8–14; Markides, C., and Geroski, P. 2005. Fast Second: How Smart Companies Bypass Radical Innovation to Enter and Dominate New Markets. Markets . San Francisco: Jossey-Bass.

T ECHNOLOGY  S- C URVES Technology S-curves are graphical representations of the development of a new technology. They compare some measure of performance (for example, speed, cost, or capacity) with some measure of effort (for instance, R&D dollars, or person-hours devoted to research). These graphs are called S-curves because the relationship between effort and performance is typically S-shaped. Initially, performance improvements per unit of  effort are small because there are many things that you need to learn before you can improve the performance of new technologies significantly. Moreover, when you first begin to develop a new technology, you probably don’t understand the key drivers of performance. Consequently, your efforts to improve the technology involve much trial and error, with many dead ends, before you figure out how to

generate performance improvements. As a result, initially you get very little performance improvement for each unit of effort. But once you identify the key drivers of performance, rapid improvement tends to follow and you see a big performance improvement for each unit of effort. 13 For instance, the first airplanes required a significant investment in research on wing and engine design to achieve a few additional feet of flight. However, once the basic design of the airplane was identified, the performance of aircraft increased dramatically. While these improvements to the technology can continue at a rapid pace for a while, at some point the technology reaches its physical limits, and diminishing returns begin to set in, leading the curve to take on the upper portion of the S-shape. That is, you start to need a lot of effort again to get each unit increase in effort. For instance, conventional semiconductors have been improved over the past three decades by doubling the number of transistors per square inch of silicon every couple of years. However, the dimensions of semiconductors have now shrunk to the point that the photographic techniques that semiconductor manufacturers employ to produce them can no longer be used to make them smaller, causing the law of diminishing returns to kick in. (Figure 5 shows the technology S-curve for Intel’s semiconductors.)

22

 

Technology Evolution

FIGURE 5 45,000,000

S-Curves in Microprocessors

40,000,000   s   r

35,000,000

   t   o   s    i   s   n   a   r    T    f   o   r   e    b   m   u    N

30,000,000 25,000,000 20,000,000 15,000,000 10,000,000 5,000,000 0 0

5,000

10,000

15,000

20,000

25,000

Cumulative Expenditure on R&D (millions)

Intel’s development different generations of in microprocessors shows thetechnology traditionalinitially S-shaped curve of technology of development; investment the development of the leads to a small improvement in performance—in this case the number of transistors on a chip— followed by rapid acceleration in performance and, ultimately, diminishing returns. Source: Adapted from data contained in Schilling, M. 2005. Strategic Management of Technological Innovation. New York: McGraw Hill; www.icknowledge.com/trends/uproc.html.

You need to understand how technology advances along an S-curve to formulate an effective technology strategy because this pattern of technical advance affects who innovates successfully in an industry industry.. Technological improvements along an S-curve tend to be incremental, building on prior developments, and taking place within an existing paradigm. Consequently, established firms, with experience operating in the industry, tend to be the ones that make these types of technological advances. First, they have existing technical, market, and organizational capabilities, which they can use to make improvements to the underlying technology with which they are working.14 Second, they have an existing customer base that provides them with information about market needs, which they can use to introduce new versions of their products and services that appeal to customers. 15 Third, they have access to internal cash flow to invest in the improvement of the technology without having to raise money from external investors. Because external investors have less information about the development of new products than the companies developing those products, they demand a premium for financing new product development, relative to the cost of using internal cash flow, putting new companies at a financial cost disadvantage in developing new products and services.

Shifting S-Curves When an existing technology reaches the point of diminishing returns, it becomes difficult to achieve very marked performance improvement by making incremental advances to it. At this time, a new technology is often developed to challenge the existing technology technology..16

23

 

Technology Evolution

The new technology typically differs from the old one in certain important ways. When it is first introduced, the new technology is usually inferior to existing technology on key dimensions, such as quality, performance, or reliability. 17 For example, the quality of printed pictures of the first digital cameras (which cost about $1,000) was inferior to that of disposable 35 millimeter cameras (which cost about $10). However, because the new technology is a radical change from the existing technology (it is based on fundamentally different technical principles), it has greater potential for performance improvement imp rovement than the existing technology technol ogy.. For instance, in the 1990s, the rate of improvement of digital camera technology was greater than that of chemical film technology. As a result, the performance of digital film technology reached a point where it surpassed the chemical film alternative. The introduction of new technologies can be represented graphically by a shift in the S-curve. As Figure 6 shows, traditional traditional telecommunications telecommunications switching technology has reached the point of diminishing returns, leading leadi ng an alternative called Voiceover-Internet overInternet protocol protocol (VOIP) to emerge. VOIP is less expensive than traditional telephone technology because the voice messages are divided into packets of data, which are sent over the most efficient route to the receiver and reassembled at the other end, rather than being sent directly over dedicated lines. 18 As Figure 6 indicates, VOIP began with a lower level level of performance than than traditional telecommunications technology improved at aithigher than traditional telecommunications technologybut to then ultimately surpass on therate performance dimensions that customers care about. Once a new technology improves to the point where it surpasses the old technology on the dimensions that customers value, customers shift to the new technology in large numbers. The new technology then takes off, and the old technology goes into decline. 19 FIGURE 6

Shifting S-Curve in Telecommunications

Improvements in Traditional Telecom Switching   e   c   n   a   m   r   o    f   r   e    P

VOIP

Effort

This figure shows a shifting S-curve in telecommunications; initially the performance VOIP was inferior to the technology that it replaced, but its rate of improvement has been higher,ofallowing it to overtake the performance of the traditional telecom switching technology. technology.

24

 

Technology Evolution

Because technologies eventually reach a point of diminishing returns, companies using a particular technology often investigate alternative technologies that can provide greater performance improvement in the future. For instance, decreasing returns in microchip density have led Hewlett-Packard to use nanotechnology to develop tiny junctions of titanium and platinum wires a few atoms wide.20 The goal of this effort is to fit more transistors on the chip, in this case increasing the number of transistors eight times, while reducing energy use. 21 Because these approaches to making semiconductors are fundamentally different from current approaches to producing them, they are more likely to permit higher transistor density than incremental improvements to existing technology.

Who Shifts the S-Curve? Even though incumbent firms typically investigate new technologies when the core technology with which they are working reaches the point of diminishing returns, new entrants tend to be the ones to introduce products and services based on new technologies. Even when they have developed the new technologies, incumbent firms tend to persist with investments in their current technologies,22 and rarely shift to newer technologies until the new technologies challenge their ability to serve their core customers with their core products.23 For example, AT&T AT&T was slow to shift to VOIP even though its researchers invented the voice compression technology that made Internet telephone calls possible;24 and Eastman Kodak was slow to shift to digital cameras even though researchers at that company invented the digital camera in 1976.25 So why do new entrants introduce the new technologies that shift the technology S-curve? Basically, Basically, there are six reasons: 1. Incumbents have no no incentive to introduce introduce the new technology: technology: The performance of new technology is typically inferior to the performance of existing technology when it is first introduced, giving incumbent firms the option of changing to a technology that will produce less income (sometimes no income) or making an incremental change to their existing technology. technology.26 For instance, the major semiconductor firms are not switching from dynamic random access memory (DRAM) chips to magnetic random access memory chips (MRAM) even though the latter permit data to be stored longer. Why?

Because the first generation of MRAM chips cost $25 each, and store four megabits of data, while DRAM chips cost $5 each, and store 512 megabits of  data. Thus, the storage capacity and price of MRAM chips is not yet competitive with DRAM chips.27 2. Incumbents have investments in existing technology: Incumbents have to write-off their investments in the old technology to adopt the new technology, and the size of these write-offs can be very large when new technologies require fundamentally different assets to be produced. For example, many of the incumbent telephone companies did not want want to launch VOIP phone service  because they had large investments in copper phone lines li nes that Internet phones 28 would make obsolete. 3. Products based based on the new technology technology cannibalize cannibalize incumbents’ incumbents’ sales:29 (Cannibalization occurs when a firm sells a new product that takes sales from one or more of its existing products. 30) If a product based on a new technology completely cannibalizes existing sales, then introducing a product based on a new technology generates additional costs, but no extra revenue, making it financially unattractive.31 For example, the makers of copper cable for

25

 

Technology Evolution

TABLE 1

The Failure to Foresee the Value of New Technologies Technological evolution is hard to forecast; even very knowledgeable people fail to see the value of new technologies in their fields. TECHNOLOGY

QUOTE

PERSON

Phonograph

Thcommercial e phonograpvalue. h . . . is not of any

Thomas Alva Edison, phonograph, 1880 inventor of the

Airplane

Heavier-than-air fl flying ma machines are impossible.

Lord Kelvin, British mathematician, physicist, and President of the British Royal Society, Society, 1895

Computer

I think there is a world market for about five computers.

Thomas J. Watson, Chairman of IBM, 1943

Pers Pe rsona onall Com Compu puter ter

Theree is Ther is no no rea reaso son n for for an any y in indiv divid idua uall to have a computer in their home.

Ken Olson, President of Digital Equipment Corporation, 1977

Telephone

The “telephone” has too many shortcomings to be seriously considered as a means of communication. The device is inherently of no value to us.

Western Union internal memo, 1876

Television

Television won’t be able to hold on to any market it captures after the first six months. People will soon get tired of  staring at a plywood box every night.

Darryl Zanuck, Head of 20th Century Fox Films, 1946

Movies wi with So Sound Who th the he hell wa wants to to he hear ac actors ta talk? Harry M. M. Wa Warner, 19 1927 P., and V. V. Mavaddat. 2000. Scenario planning for disruptive technologies. In G. Day and Sources: Adapted from Schoemaker, P., P. Shoemaker (eds. (eds.) Wharton on Emerging Technologies Technologies.. New York: John Wiley; and Mohr, J., S. Sengupta, and S. Slater. 2005.  Marketing of High Technology Technology Products and Innovations (2nd edition). Upper Saddle River, NJ: Prentice Hall.

telecommunications resisted the shift to fiber optic cable because the latter was a substitute for the former and selling fiber optic cable would completely cannibalize their sales to telecommunications companies. 4. Managers at incumbent firms firms do not see the new technology as a threat (see Table 1): To maintain focus on their thei r core activities, established establi shed firms create routines that lead managers to filter out information about alternatives to their core technologies, which leads le ads them to view new technologies negatively.32 For example, Epson did not shift to the production of ink-jet printers after HewlettPackard themalternative because itsfor management believed were notintroduced a cost-effective most consumers, andthat thatink-jet ink-jetprinters print33 ers would not challenge the dot-matrix printer business. 5. Incumbent firms can improve the performance of their old technologies: technologies:34 For instance, Jim Utterback, a professor of technology management at MIT, has shown that the producers of mechanical typewriters, sailing ships, mechanical watches, and the telegraph all improved those products significantly in response to the introduction of the new radical technologies that ultimately replaced them. 35 6. Incumbe Incumbent nt firms face organizational organizational obstacles obstacles to changing their core technologies: When a new technology comes along, organizations often have to change their structures to fit the new technology. This causes fighting within the organization because some managers have to lose political power and influence. 36 These managers resist the changes, even if they are necessary for the organization to survive. For example, Polaroid failed in its efforts to exploit the shift to digital camera technology because its managers resisted making the people in charge of  electronic digital signal processing, software, and storage technologies more central to the organization than the people in charge of optics and film technology.

26

 

Technology Evolution

Using S-Curves as a Management Tool Tool Richard Foster, Foster, a management consultant who has written extensively about technology S-curves, suggests that you can use them as a tool to predict when your company should invest in the development of a new technology. technology. For example, Hewlett-Pack Hewlett-Packard ard used the technology S-curve to determine that it should invest in the development of  a line of photo printers back in 1995, when traditional film cameras still were dominant, and digital cameras were very expensive.37 By plotting data on the relationship  between the amount of investment that has been made in a technology against data on key performance indicators, you can determine where a current technology lies on the S-curve, and whether it has reached the point of diminishing returns.38 If plots of  the technology S-curve reveal that your company’s core technology has reached the point of diminishing returns, then you should prepare for a technological shift by investing in the development of alternative technologies. However,, there are several important limitations However limitatio ns to technology S-curves as a tool to help formulate your firm’s technology strategy. First, plots of technology S-curves will tell you that your company needs to change from the old technology to the new technology,, but not when your company should make that change. Few technologies technology have known performance limitations, making it difficult to determine the amount of  improvement to the old technology that is possible once the new one has been introduced. Take, for example, the internal combustion engine and the fuel cell. It is not possible to predict when companies should switch from making vehicles with internal combustion engines to ones with fuel cell engines because we do not know when the performance of the fuel cell will exceed the performance of the internal combustion engine. The inability of S-curves to predict when you should shift from an old technology to a new one is an important limitation because shifting too early will saddle your company with a new technology that is inferior to the old one, while switching too late will allow other firms to move up the learning curve and position their product or service as the dominant design or technical standard in the industry. Moreover,, in many cases, several things have to happen before a new technology Moreover can “take off,” and it is hard to know when that confluence of events will occur. Take, Take, for example, digital camera technology. The replacement of traditional film cameras with digital cameras did not take off until personal computer manufacturers had developed computers with the processing and storage capability for basic imaging, and Second, digital cameras could to beswitch designed usetechnology computer storage and processing. your decision to atonew shouldn’t just depend on the old technology having reached the point of diminishing returns. It should also depend on your identification of a need for a product based on the new technology, your ability to get customers to switch to products based on the new technology, technology, your company’s capability to produce products based on the new technology, and the expected rate of diffusion of products based on the new technology. 39 Existing technologies will often reach a point of diminishing returns long before you can offer a viable alternative based on a new technology. For example, the shift to the technology of e-books has been very slow even though improvements to paper books long ago reached the point of diminishing marginal returns. It can be particularly difficult to get industrial customers to switch to products  based on a new technology because product improvements based on existing technology are much easier to integrate into their product development and production activities than are new products based on fundamentally new technology. If your customers need to change their infrastructure or production and product development processes to adopt products based on a new technology, technology, they are going to resist

27

 

Technology Evolution

that change. As a result, you will risk losing your existing customers by making the shift to a new technology technolo gy instead of improving your old technology.40 Third, switching to the new technology may not make sense because the new technology is initially inferior to existing alternatives and, therefore, cannot be used to serve the mainstream of the market. To use new technology successfully, you first have to focus on segments of the market that may be less profitable, while investing in the further development of the new technology to get it to the point where it is competitive with existing technologies in mainstrea mainstream m markets. 41 For example, the companies that exploited exploited VOIP telephone service first had to target niche markets markets and improve the technology before they could offer good enough call quality to serve the mainstream of the telecommunications market. The need to focus on lower margin niche markets for an indeterminate amount of time might make the shift to the new technology too costly for your company. Fourth, as the manager of an incumbent firm, you might be able to deal with decreasing marginal returns to your technology’s performance without immediately changing to a new technology. You can improve your product architecture, enhance components,42 identify new applications for your products, or even fight the adoption of new technology with legal action or public relations efforts, to maintain, or even increase, your sales. As a result, you might be better off waiting to shift your company to the new until afterover it offers better performance than the existing technology. Yourtechnology company’s control assets in manufacturing, marketing, and distribution, and your brand name, could allow you to catch up to new firms that were the first to adopt the new technology. For instance, AT&T is betting on the value of its brand name with consumers to allow it to catch up to Packet8 and Voice Pulse, VOIP start-ups that have seen rapid growth growth in their number of customers customers in 43 the last few years.  Key Points

• New technologies evolve in an evolutionary manner, along technological trajectories, constrained by the paradigms within which scientists and engineers work. • Incremental change involves small changes to a technology within an existing paradigm; radical change involves large changes to a technology and a break with the existing paradigm. • The development of new technologies tends to follow an S-shaped pattern: In the initial period, large amounts of effort yield small performance improvements; in the intermediate period, small amounts of effort yield large performance improvements; and in the final period, large amounts of effort again yield small performance improvements. • New entrants typically shift the S-curve because the new technologies start with lesser performance than existing alternatives, require existing assets to be written off, cannibalize the sales of existing products, and are often dismissed by managers of incumbent firms, who face significant organizational pressures to maintain existing technology and who can improve their existing technology. • Incumbent firms can use technology S-curves as a management tool to predict when to invest in a radical new technology; however, the use of S-curves for this purpose faces important limitations, including the inability to identify when to switch technologies, the failure to incorporate all of the factors that matter to the decision to switch, the need for adopters of the new technology to focus on niche markets before tackling the mainstream of the market, and the existence of alternative ways to respond to the introduction of new technology.

28

 

Technology Evolution

T HE  A BERNATHY- U TTERBACK   M ODEL The late Bill Abernathy, a professor at Harvard Business School, and Jim Utterback, a professor at the Sloan School of Management at MIT, proposed a model of technology evolution that expands upon the basic ideas of the technology According to Abernathy and Utterback, technology evolves through periods S-curve. of incremental innovation, interrupted by periods of radical innovation. The development devel opment of  a radical innovation leads to a fluid phase in an industry, during which time many firms enter and compete on the basis of different product designs. Eventually, the firms in the industry converge on a dominant design, which results in the specific phase, during which time only incremental innovation occurs. After a while, the cycle repeats with the development of a new radical innovation, which introduces a new fluid phase.

The Nature of Innovation and Competition This evolutionary cycle that professors Abernathy and Utterback described affects the nature of innovation and firm competition. To see how the evolutionary cycle affects the nature of innovation, you need to understand the difference between product and process innovation. When technological innovation involves the creation of new goods and services sold to customers, product innovation is the term used to describe it. The DVD player is an example of a product innovation  becausee technical  becaus techn ical knowled kn owledge ge was used us ed to create crea te somethin som ething g new that tha t could be sold to customers. When technological innovation involves problem solving that improves the method of creating or delivering a product or service, it is called process innovation. For example, Boeing engaged in process innovation when its engineers figured out how to make the wings of the new Dreamliner aircraft out of a composite material in place of aluminum to make the plane lighter and more fuel efficient. The innovation in the materials used to make the wing was a process innovation because it did not lead to a new product or service purchased by customers, but, instead, led to a better way of making an existing product or service. The initial phase of an industry is called the fluid phase. This phase is a period of  high uncertainty. During this phase, markets are relatively small, customers do not have clearly defined preferences, and the diffusion of the industry’s core technology is quite limited. Companies that enter the industry during the fluid phase engage primarily in product innovation, competing on the basis of the novelty of their product designs. For instance, in the beginning of the twentieth century, when the automobile industry had not yet converged on a dominant design of a four-wheeled vehicle with an enclosed body and a steering wheel, some cars were sold with three wheels instead of four, four, with open bodies instead of closed bodies, and with joysticks instead i nstead of steering wheels. During the fluid phase, production is relatively inefficient. Initial entrants are often small and use a variety of different production processes. They engage in very little process innovation and tend to use generic inputs and production equipment. 44 The fluid phase of technology evolution comes to an end when the industry condesign45 verges a dominant , or a common way that all companies a product willon design that product. The internal combustion engine is a producing good examples of  a dominant design. For decades, all automakers have produced vehicles using this

29

 

Technology Evolution

design; the original steam and electric vehicle alternatives to this design are now long gone. (The fact that car companies are now considering making electric vehicles again, approximately one hundred years after the industry abandoned that design alternative, is evidence of the cycle of radical and incremental innovation that Abernathy and Utterback identified.) The rise of the dominant design leads to the specific phase of industry evolution. During this phase, competition shifts away from design uniqueness to production cost, as firms focus on achieving production efficiencies with the common design. 46 The standardization that comes from convergence on a dominant design allows for the introduction of interchangeable parts, and an agreed-upon format for components,47  both of which permit investmen investmentt in speciali specialized zed long-liv long-lived ed assets, such as 48 manufacturing equipment, and efficient volume production. As a result, production shifts to large firms that can exploit economies of scale, allowing the cost of production and prices to fall. 49 Ironically, the diffusion of new products tends to take off during the specific phase, when customer preferences stabilize and product variety is reduced. Most customers will not adopt new products until there are clear features and metrics on which those products can be evaluated, and these features and metrics tend not to be identified until after a dominant design emerges. When firms innovate during the specific phase,tothey tenddramatically to focus on process innovation, leading the rate of product innovation decline during 50 this phase of industry evolution. (Figu (Figure re 7 shows how the timing timing of product and process innovation changes over the industry life cycle.) The personal computer industry provides a good example of how the dominant type of innovation changes as an industry shifts from the fluid to the specific phase. In the early days of the personal computer business, many different companies competed on the basis of different product designs, with companies

FIGURE 7

The Timing of  Product and Process Innovation

  n

Product Innovation

   i    t   o   a   v   o   n   n    I   r   o    j   a    M    f   o   e    t   a    R

Process Innovation

Fluid Phase

Transitional Phase

Specific Phase

The Abernathy-Utterback model holds that the level of product innovation tends to decrease over time, while the level of process innovation tends to peak in the transitional phase of industry evolution. Source: Utterback, J. 1994. The Dynamics of Innovation. Innovation. Boston: Harvard Business School Press.

30

 

Technology Evolution

offering a variety of disk formats, microprocessors, and operating systems. 51 Then a dominant design emerged and personal computers became standardized. As a result, companies in the industry reduced spending on product innovation and focused their attention on efforts to reduce manufacturing, distribution, and marketing costs.

New Firm Performance The Abernathy-Utterback model provides insight into when new and established firms are most successful in technology-intensive industries. In general, the fluid phase of an industry is the most favorable to new firms, while the specific phase is most favorable to incumbent firms. Several factors account for this pattern. 1. Before the establishment of a dominant design, new firms can operate without adopting the same product design as more experienced firms. However, once a dominant design has been adopted, new firms must adhere to the standard product design in the industry. Because established firms have greater experience working with this design than new firms do, new firms are disadvantaged when all firms have to use the same product designs.52 2. Before the establishment of a dominant design, firms operate on a small scale to minimize technical uncertainty uncertainty..53 Moreover Moreover,, they tend to have nonhierarchical organization structures because these structures facilitate product design and development. However, once a dominant design has been established industrywide, the production process becomes standardized, and competition shifts to production efficiency and economies of scale.54 To support this need for efficiency,, organization structures become more hierarchical and bureaucratic, ciency which favor large, established businesses. 3. Before a dominant design emerges, learning curves are weak, allowing new firms to enter without operating at a severe competitive disadvantage. However, after a dominant design emerges, learning curves become more pronounced. Because efficient manufacturing, effective selling, and careful response to customer complaints all involve learning by doing,

established firms, which have more operating experience than new firms, perform better.

Number of Firms in the Industry The number of firms in an industry also tends to change over the different phases of  industry evolution. In almost all cases, during the fluid phase, there is a high rate of  new firm entry and a low rate of new ne w firm exit, leading to a large increase in the num ber of firms in the industry. industry.55 For example, during the fluid phase of the automobile industry (from 1895 to 1923), the number of automobile manufacturers increased from 0 to 75. Once the industry converges on a dominant design, a shakeout ensues as the industry consolidates around the small number of firms that are able to develop products that fit the dominant design. 56 As Table Table 2 shows, the magnitude magnitude of the shakeout is quite dramatic in most industries, with the drop in the number of firms from the peak level averaging 52 percent, and going as high as 87 percent. 57

31

    

Technology Evolution

TABLE 2

The Severity of the Shakeout This table shows the results of a study measuring the number of firms that exited several industries and the rate of  decline to a stable number of firms; it shows that all industries experience a shakeout of firms, but the severity of the shakeout varies substantially across industries. PRODUCT

PEAK NUM UMBER BER OF FIRMS

DECLI ECLINE NE TO STABLE LEVEL

DDT

38

87%

5

Streptomycin

13

85%

2

Penicillin

30

80%

6

Automobile tires

NUMB UMBER ER OF FIRMS REMAINING

275

77%

63

Radio transmitters Saccharin

76 39

72% 72%

7 11

Electric blankets

17

65%

6

Freezers

61

62%

23

Phonograph records

49

61%

19

Windshield wipers Electric shavers

51 32

59% 56%

21 14

Photocopy machines

43

53%

20

Adding machines

55

51%

27

Fluorescent lamps

34

41%

20

Piezo crystals Outboard motors

45 21

38% 38%

28 15

Polariscopes

16

38%

16

Cryogenic tanks

84

35%

55

 Jet engines

29

31%

20

Cathode ray tubes Zippers

39 49

28% 18%

28 40

114

4%

109

Shampoo Average

52%

Source: Adapted from Klepper, S., and E. Graddy. 1990. The evolution of new industries and the determinants of market structure. Rand Journal of Economics, 21(1): 32.

Limitations to the Model While the Abernathy-Utterback Abernathy-Utterback model is quite useful, it does not predict the patterns of  industry evolution equally well in all manufact manufacturing uring industries. It appears to work best in assembled products in which customer tastes are homogenous, and worst in nonassembled products, like rayon or glass, or in products that involve nonassembled components, like integrated circuits.58 With these technologies, firms engage in less product innovation innovation and more process innovation, altering how innovation and competition change as industries transition from the fluid phase to the specific phase of industry evolution. Perhaps more importantly, the model is not very effective in predicting patterns of industry evolution in service businesses, a point discussed in more detail next.  Key Points

• According Abernathy and Utterback, technology evolves through of  incrementaltochange, interrupted by radical change, which affects bothperiods the nature of innovation and firm competition.

32

 

Technology Evolution

• During the fluid phase, markets are uncertain, customer preferences are illdefined, diffusion of the technology is limited, firms are small, and most innovation takes the form of new products. • The fluid phase ends with convergence on a dominant design, which ushers in the specific phase, during which time firms shift to process innovation, produce standardized products, and strive for production efficiencies based on scale economies and interchangeable parts. • New firms perform best in the fluid phase of industry evolution because they are better than established firms at product innovation and worse than established firms at efficient production based on scale economies. • During the specific phase, a shakeout typically occurs, with approximately half  of the firms exiting the industry; those firms least able to fit their operations to the dominant design tend to exit. • The Abernathy-Utterback model holds best in assembled manufacturing in which consumers have homogenous tastes; it holds less well in nonassembled manufacturing and manufactured products based on nonassembled components; and it does not hold in services.

GETTING DOWN TO BUSINESS

The History of Electric Ve Vehicles hicles59

While most people think that the use of electricity to power hybrid vehicles is a novel idea, electric-powered vehicles actually were once the most common type of  automobiles in the United States. In fact, at the turn of  the twentieth century, most experts predicted that the electric engine would become the dominant design in automobiles, largely because of their greater range and power. Of course, that didn’t happen. So why did the internal combustion engine become the dominant design for automobiles, and the electric engine fall by the wayside for close to 100 years?

The initial advantage of gasoline-powered cars for touring was reinforced by increasing returns to adoption. Because more people selected gasoline powered vehicles, the fueling stations that developed as a complement to automobiles focused on gasoline, not electricity, and the mechanics that repaired vehicles for others concentrated their skills on the more popular type of vehicle. These two trends reinforced the initial tendency of people to choose gasoline-powered vehicles, and the internal combustion engine became the dominant design. Once this happened, most of the

Technical and social factors, combined with increas- producers of electric vehicles closed down. ing returns to adoption, were largely responsible for the Once the dominant design emerged, the automobile convergence on internal combustion as the dominant industry changed dramatically. The companies founded design in automobile engines. Even though electric vehi- to provide gasoline-powered vehicles did much better cles had the technical superiority of range and power, than the ones founded to offer electric-powered vehicles they had an important weakness. They weren’t very  because  because it was hard hard for for the the makers makers of elect electric ric vehicle vehicless to good for touring, and the main purpose for automobiles a utomobiles transition to another core technology. As a result, many at the turn of the twentieth century was fun. (Horse- of the electric vehicle makers exited the industry industry,, and the powered wagons still did most of the actual work). The gasoline-powered vehicle makers grew into large, hierproblem that electric vehicles faced for touring was that archical organizations that mass produced cars and took there was no way to recharge batteries. So if you got lost advantage of economies of scale. New firms stopped or were slow getting back home, you could run out of  entering the industry in large numbers, and the few  battery power and have no way to refuel. In contrast, large companies in the industry competed amongst with a gasoline-powered car, you could just pack some themselves, largely on the basis of process innovation, extra gas in a tank and go off with little worry that you efficiency, and marketing. And for many decades, no would run out of fuel. manufacturer offered a car powered by electricity. electricity.

33

 

Technology Evolution

M ODIFICATIONS TO THE A BERNATHY-U TTERBACK  M ODEL Although the Abernathy-Utterback Abernathy-Utterback model has proven to be quite useful in explaining the evolution of technology technology,, researchers have identified four important modifications that you need to understand: Barras’s reverse product cycle theory; Tushman’s Tushman’s model of competence-destroying and competence-enhancing innovation; Christensen’s model of value networks and disruptive innovation; and Henderson and Clark’s model of architectu architectural ral innovation.

Reverse Product Cycle Theory After Abernathy and Utterback introduced their model, researchers wondered if  it applied to service businesses as well as to the manufacturing businesses on which it had been developed and tested. When Dr. Richard Barras, a London based  base d researche rese archer, r, looked looke d at this t his quest q uestion, ion, he conclud co ncluded ed that th at the answe answerr was no. A different differ ent model operated in service industries, which he called the “reverse product cycle.” According to Barras, service industries typically adopt new technologies that are first developed in a goods industry.60 For instance, the insurance industry adopted computer technology that was developed by the manufacturers of mainframe computers. This adoption of technology from a goods industry leads to the first stage of  the reverse product cycle. The adopted technology is employed to make existing services more efficient, thereby reducing costs. 61 These initial innovations are typically incremental improvements.62 For example, the insurance industry first used mainframe computer technology to create computerized records of insurance policies. 63 In the second stage of the reverse product cycle, the new technology is used to make the service more effective. The innovations that are introduced at this stage tend to focus on changing the processes used to serve customers, thereby enhancing quality.64 For example, in the 1980s, insurance companies began to use information technology to offer better claims service, with greater hours and more alternatives. 65 In the third stage of the reverse product cycle, the technology is used to create new services, making changes at this stage radical.66 For example, in the 1990s, insurance companies used network computing to offer online quotation services to their customers.67 Comparing the reverse product cycle model with the Abernathy-Utterback model, you can see that the timing of radical and incremental change is different in manufacturing and service businesses: Manufacturing industries start with radical innovation and move to incrementa incrementall innovation, while service industries start with incremental innovation and move to radical innovation. 68 As a result, the timing of advantage for new and established firms is different in manufacturing and services. In manufacturing, new firms are most advantaged when new technology is first developed and a dominant design has not yet emerged. By contrast, in services, established firms initially benefit the most from the new technology because it reduces the cost of delivering existing services. It is only later on, have whenbeen the technology the creation new services,shifts and barriers to entry reduced bypermits the decline in costs, of that advantage to new 69 companies.

34

 

Technology Evolution

Competence-Enhancing and Competence-Destroying Innovation In the years after Abernathy and Utterback published their model, researchers noticed an important puzzle that the model could not explain. In some cases, incum bent firms had little trouble transitioning to new radical technologies. te chnologies. For instance, i nstance, General Electric was able to shift from making medical diagnostic devices based on X-ray technology to those based on CAT scans to those based on MRIs, even though all three technologies are fundamentally different.70 This puzzle led Professor Michael Tushman of Harvard Business School and his students to modify the Abernathy-Utterback model to explain why incumbent firms were sometimes able to transition to radical new technologies and other times were not. Professor Tushman and his colleagues explained that radical new technology does not always undermine the capabilities of incumbent firms because sometimes it can be competence-enhan competen ce-enhancing. cing. A radical technology is competence-enhancing if it makes use of existing knowledge, skills, abilities, structure, design, production processes, and plant and equipment; whereas, it is competence-destroying if it undermines the usefulness of these things. For example, the shift from vacuum tubes to integrated circuits in computers destroyed existing competencies because it rendered obsolete the expertise that firmsexplained, had developed in vacuum tubes. As Tushman established firms are able to transition to a radical technology when that technology is competence-enhancing but fail to do so when it is competence-destroying. When radical technological change is competenceenhancing, incumbent firms invest in its development because they have an incentive i ncentive to do so. Moreover, they have the capabilities to develop that technology successfully. succe ssfully. For example, incumbent firms were able to transition to the use of turbofans in jet engines because this technology was competence-enhancing and drew heavily on the skills that jet engine manufacturers already had.71 However, incumbent firms cannot transition easily to competence-destroying radical technological change, often leaving that type of technology to firms outside the industry, or, in many cases, to start-up companies. Even when incumbent firms invent competence-destroying technologies, they rarely introduce them. Competence-destroying radical technological change tends to be financially costly for incumbent firms, giving them little incentive to invest in their development. 72 Moreover, their existing capabilities do not help them make use of those technologies, and, in many cases, actually hinder their ability to develop them by requiring incumbent firms to unlearn their old ways of doing things.73 For instance, many incumbent firms in the printing industry were unable to transition to digital printing because this technology was competence-destroying and required the firms to develop skills in database systems, which they did not have, to manage the printing process.

Architectural Innovation Rebecca Henderson, a faculty member at the Sloan School at MIT, and Kim Clark, a faculty member at the Harvard Business School, offered a different modification to the Abernathy-Utterback model from that offered by Tushman and his students. Henderson and Clark noticed that incumbent firms often failed to manage the transition from technology another, even when those technologies were not radical. Xerox, forone example, had to severe problems developing small copiers based on the underlying plain-paper copier technology that it had pioneered.

35

 

Technology Evolution

To explain why incumbent firms were tripped up by what seemed to be rather incremental technological changes, Henderson and Clark developed a more finegrained taxonomy of innovation than the one on which the Abernathy-Utterback model is based. In this taxonomy, two additional types of innovation not mentioned  by Aberna A bernathy thy and Utte Utterback rback are prese present: nt: modular innovation and architectural innovation. A modular innovation innovation is one that changes the components components from which the innovation is created, but not the linkages between those components. For example, the digital telephone is a modular innovation because it and the analog telephone have fundamentally different components, but the same linkages between them. An architectural innovation is one that changes the linkages between the components, but leaves the components themselves intact. 74 For example, the portable DVD player is an architectural innovation. The design of its components is the same as with the desktop DVD player, but the architecture—how a built-in screen is linked to the rest of the device, how the device is powered, and so on—is different. 75 The addition of modular and architectural innovation to the Abernathy and Utterback taxonomy leads to the two-by-two matrix shown in Figure 8. Henderson and Clark explain that incumbent firms often fail in the face of architectural innovation for three reasons. First, incumbent firms often lack the right external linkages to gather information about a new technology architecture emerging in an industry. Because organizations adopt external ties that are appropriate for their current activities, their external linkages facilitate the exchange of information with suppliers and customers about their current product architecture and discourage the exchange of information about other possible architectures. 76 Second, incumbent firms often lack the capacity to recognize the value of information about architectural innovation that is presented to them. Companies need to

FIGURE 8 Core Concepts Reinforced

Core Concepts Overturned

Linkages Unchanged

Incremental

Modular

Linkages Changed

Architectural

Radical

Henderson and Clark’s Taxonomy of Innovations

Henderson and Clark developed a taxonomy of innovation based on the degree of change to core concepts and the linkages between them; they found fo und that established firms have many problems responding to architectural innovation. Source: Adapted from Henderson, R., and K. Clark. 1990. Architectural innovation: the reconfiguration of  existing product technologies and the failure of existing firms. Administ firms.  Administrative rative Science Science Quarterly, Quarterly, 35(1): 9–30.

36

 

Technology Evolution

have existing, related knowledge to recognize the value of information about architectural innovation. However, once an industry has converged on a dominant design, firms accept the product architecture as given, and focus their attention on improving components or technological processes. As a result, they stop making investments in alternative product architectures, which keeps them from developing the background necessary to absorb external information about new architectures.77 Third, even when incumbent firms recognize the value of architectural innovation, they often have difficulty making use of it because adopting an architectural innovation typically requires a company to restructure. Organizations typically align their structures to their product architecture (see Figure 9). For instance, automobile manufacturers typically develop a steering gear organization composed of a steering column group, a rack-and-pinion group, a power steering stee ring group, and a tie rod group  because this paral parallels lels the archit architecture ecture of a steer steering ing column column..78 Thus, to change its product architecture, a company must also change its supporting organizational structure, changing who reports to whom and who talks to whom inside the organization. This, of course, is a difficult and costly activity activity..

Disruptive Technology and Value Networks Clayton Christensen, a Harvard Business School professor, was puzzled by a different problem with the Abernathy-Utterback model than that which puzzled Tushman, and Henderson and Clark. Christensen noticed that the incumbent firms that were unable to adopt the radical new technologies were often the very firms that invented them. This pattern suggested that a lack of technological capability could not be the explanation of the firms’ failure to transition to the new ne w technology. technology. Christensen believed that the source of the problem was the willingness of a company’s customers to adopt products and services based on the new technology, and

FIGURE 9 Product Architecture

Organizational Structure Mirrors Product Architecture

Organizational Structure

1

2

1

2

3

4

3

4

5

5

One reason why architectural innovation is difficult for incumbent firms is i s that organizational structure evolves to mirror product architecture, architecture, requiring the organizationa organizationall structure to change for the product architecture to change. Source: Adapted from Christensen, C. 1997. The Innovator’s Dilemma. Dilemma. Boston: Harvard Business School Press.

37

 

Technology Evolution

not on the company’s technical capabilities. He found that engineers at incumbent firms often developed radical, or what he called disruptive, technologies. When the companies produced new products products based on the new technologies and showed them to their existing customers, the customers rejected them because they were typically inferior to existing technology on some dimension that was important to the customers.79 For example, Teradyne’s semiconductor testing-equipment company was unable to adopt complementary metal-oxide semiconductor (CMOS) chips and Windows-based Win dows-based software in their testing equipment because their primary customers were worried about the reliability of CMOS chips and software compatibility.80 As a result, the primary customers of incumbent firms would tell the marketing personnel to make incremental improvements products based on their old technology. technology.81 The companies would then focus on these incremental improvements to ensure that they satisfied their mainstream customers on whom their sales and profits depended. Often the employees that developed the new technologies did not agree with the decision to focus on incremental improvements, and quit to start new companies to exploit the new technology.82 These employees-turned-entrepreneurs had no more success serving the mainstream customers of their former employers than their former employers themselves, and had to turn to new markets to find customers interested in buying their products. The new companies often found these customers in 83

segments of the market that had not been served well previously. For example, take the case of a radical new technology used in memory chips, MRAM, which combines magnetized material and silicon to allow data to be stored magnetically, rather than electronically. Magnetic storage allows data to be stored indefinitely, which is not possible with DRAM chips, and also allows data to be accessed more quickly. Because MRAM chips are much more expensive than DRAM chips, the producers of cell phones and personal computers, which are big users of  DRAM chips, are not interested in the product. However, Freescale Semiconductor, the company selling MRAM chips, has found a market in the makers of home security devices, networking devices, and server systems, which are not adequately served by DRAM chips because they need permanent storage and rapid data access.84 In Christensen’s model, the new firms provide products for these niche market segments, which are often the least attractive segments of the market, because they have no choice; their products lack the features necessary to satisfy the mainstream of the market. Incumbent firms often cede the niche markets to the new firms, and focus on the mainstream of the market, because the niches are not worth serving if  the incumbent firms have to compete to serve them. 85 Over time, the new firms invest in the development of the new technology. Because the potential for improvement of the new technology is higher than the potential for improvement of the old technology technology (see the S-curves in Figure Figure 6 again if  you do not realize why), the new firms are eventually able to produce products that are better than those of the incumbent firms on the dimensions that matter to mainstream customers. The new entrants then target the mainstream of the market, marke t, which is more attractive attractive than the segments that the the new entrants had initially initially chosen to serve. Because the new entrants now can provide mainstream customers with the product attributes that they want at a lower price, the new entrants now can take customers away from the incumbent firms. Take, for example, the history of computer workstations. Initially, workstations  built around the microprocessor did not have enough computing power to meet the needs minicomputer users. As a result, companies like Sun Microsystems Siliconof Graphics went after different segments of the computing market—ones and that were not terribly attractive. Because minicomputer makers like Wang and Digital

38

 

Technology Evolution

Equipment Corporation found those segments to be unattractive, they accommodated the workstation companies by ceding them the segments. Because the rate of  increase in the computing power of the microprocessor was faster than that of the minicomputer, the workstation manufacturers were able to expand from their initial market niche and take the minicomputer market away from minicomputer manufacturers. The end result was the demise of companies such as Wang and Digital Equipment Corporation.86 A similar pattern can be observed observed in minimills, which used a technology for making steel from scrap to compete with traditional integrated integrated mills. As Figure 10 shows, the minimills entered with lower quality steel than the integrated steel mills. At first the steel’s surface quality was only good enough to take customers for rebar, a type of  steel used to reinforce concrete. However, over time the minimills improved the surface quality of their steel to the point where it was better than that of the integrated steel mills and took the steel mills’ primary automotive and appliance customers.87 Christensen’s research research provides important implications for your technology strategy,, whether you are managing a new entrant or an incumbent firm. If your firm strategy is a new entrant, his research shows that it will perform better if you target a new or underserved segment of the market with a disruptive innovation, rather than targeting an already satisfied segment of the market with that innovation. If your technology allows your customers to accomplish things that they cannot accomplish with existing products, you can get a foothold in the market and get the chance to improve i mprove your core technology to the point at which you can target segments that previously preferred the older olde r technology.88 If your firm is an incumbent, Christensen’s research shows that it will perform  better if your core techno technology logy confo conforms rms to the feature featuress of the domi dominant nant desig design n  because deviation from the t he dominant design will make it harder for you to satisfy

FIGURE 10 Integrated Steel Mill Quality

Upmarket Migration in Steel

Sheet Steel Customer Minimum Structural Steel Customer Minimum

Minimill Quality

Angle Iron Customer Minimum Rebar Customer Minimum

1975

1980

1985

1990

1995

The minimills entered segments of the steel industry that had low demands for quality and expanded to segments with higher demands demands for quality over time by improving the quality of  their steel.

Source: Adapted from Christensen, C. 1997. Continuous Casting Investments at USX Corporation, Harvard Business School Teaching Note, 5-697-066, April 24.

39

 

Technology Evolution

customers once that design has emerged.89 Moreover, it indicates the perils of listening too closely to what your primary customers say. 90 Disruptive innovations typically appeal to segments of the market that are underserved, and not to the mainstream of the market.91 So, if you serve the mainstream of the market, listening to yourFurthermore, customers will you to ignore thatavoid you should adopt. as lead the manager of an disruptive incumbenttechnology firm, you can the outcomes that befell the integrated steel mills and the minicomputer manufacturers by developing a new company to exploit the disruptive technology, technology, rather than ignoring it, or trying to develop the technology within the confines of your existing organization. Creating a new business allows you to avoid the budget-driven, planning-oriented, approaches of large, established businesses that often hinder the development of disruptive technologies.92 Establishing a new business also allows you to exploit the disruptive technology on a small scale, so you can tinker with it, and get it right  before you need to serve large numbers of custome customers, rs, or spend large amounts of  money.93 And pursuing the new technology through an independent company allows you to focus on new customers rather than on existing ones, 94 which facilitates the identification of previously underserved market segments. 95 For example, when Teradyne wanted to introduce a new semiconductor-testing equipment product using CMOS chips and Windows-based software, it established an independent company to pursue customers that it had not served previously. When Teradyne’s Teradyne’s mainstream customers rejected the new product, the company still had a set of customers to whom it could sell the new product and off of whom it could drive improvements to the technology. technology.96  Key Points

• The reverse product cycle theory holds that innovation is different for services than for manufacturing; for services, process innovation precedes product innovation. • Radical new technology can be competence-enhancing or competence-destroying; incumbent firms may invent competence-destroying technologies, but they rarely introduce them because they lack the capabilities to do so. • Incumbent firms transition to competence-enhancing radical technology without much difficulty. • Architectural innovations are innovations that change how components are linked together but leave the components themselves intact. • Incumbent firms often fail in the face of architectural innovation because they lack the right external ties to learn about the new architecture and the absorptive capacity to recognize its value, and because they are unwilling or unable to restructure to make use of the new architecture. • Incumbent firms sometimes have trouble exploiting a radical technology, technology, not  because they have insufficient technical capability, capability, but because their customer  base does not want them to transition to the new technology. technology. • Mainstream customers often restrict incumbent firms to making incremental improvements to existing technology, allowing new firms to exploit disruptive technology in small, undesirable market niches. • New firms often enter undesirable niches because they have no choice, but improve their technology so that they can move upmarket to target the more desirable mainstream market. • Incumbent firms needof tothe establish or invest in new organizations if they want to exploit new disruptive technologies successfully. successfully.

40

 

Technology Evolution

DISCUSSION QUESTIONS 1. What forces explain the timing of major technologitechnological transitions? Can you predict and plan for such transitions? If so, how and when should you try? 2. How should you determine what to map on the vertical axis of an S-curve? 3. When is a technology technology S-curve analysis helpful? How should it be used? What are the limitations of  the technology S-curve as predictive tool? Is there something better? If so, what is it? 4. How are the approaches approaches to managing innovation the same and different for new and incumbent firms? Are these patterns the same in all industries? If not, how do they vary across industries? 5. Are some types of innovation better better and worse for entrepreneurs founding new firms? If so, why? If  not, why not?

6. Why do incumbent incumbent firms often fail in the face of  of  radical technological innovation? Why do these firms fail in the face of architectural innovation? Why don’t incumbent firms always fail in the face of  radical technological innovation? 7. What are the similarities and the differences in how Abernathy and Utterback, Barras, Henderson and Clark, Tushman and his students, and Christensen view technological change? 8. What is the right strategy for new and established firms in response to a disruptive technological change? Why?

KEY TERMS Architectural Innovation: An innovation that changes the way that the components of a system link together. Cannibalization: The sales of a new product that come at the expense of sales of a firm’s existing products. Competence-Destroying: A change that that undermines the usefulness of existing knowledge, skills, abilities, structure, design, production processes, and plant and equipment. Competence-Enhancing: A change that increases the usefulness of existing knowledge, skills, abilities, structure, design, production processes, and plant and equipment. Dominant Design: A common way way that all companies companies

Modular Innovation: An innovation that changes the components of a system but not the way that components are linked together together.. Paradigm: The framework within which technical problem solving occurs. Process Innovation: An innovation that enhances the way a good or service is made. Product Innovation: An innovation that leads to the creation of new goods and services sold to customers. Radical Innovation: An innovation that draws on a fundamentally new knowledge base. Specific Phase: The period of industry evolution after a dominant design emerges. Technology S-Curve: A graphical representation representation of 

producing product will designevolution their version. Fluid Phase: a given The period of technology  before a dominant design emerges. Incremental Innovation: An innovation that makes a small improvement on an existing knowledge base.

the development of a some new technology, the relationship between measure of showing performance and some measure of effort. Technology Trajectory: The path of improvement of a technology on some performance dimension.

PUTTING IDEAS

INTO PRACTICE

1. Technology S-Curves Select a technology with which you are familiar. Identify measures of performance of the technology and of the cumulative amount of effort to develop the technology over time (e.g., R&D expense, person-hours devoted to the technology, technology, etc.). Putting the performance measure on the vertical axis and the effort measure on

done graphing the technology S-curve, use the figure you created to answer the following questions: Is the technology/industry subject to “natural” technological limits? Why or why not? Has it experienced disruptions? Is it likely to do so soon? How will customer characteristics evolve as companies move along the technology S-curve? During a dis-

the horizontal axis, plot the relevant S-curve for that technology.. Then identify the point on the S-curve at technology which the technology is currently currently.. Once you are

ruption, how will the customers of the old and new technology differ?

41

 

Technology Evolution

2. Dominant Designs For an industry/technology with which you are familiar, fa miliar, identify a dominant design. Then examine the industry/technology  before and after the the dominant design emerged. emerged. Document the effect of the dominant design on

hydrogen. The only exhaust they produce is water. The facilities used to make these engines look like semiconductor factories. The key component is a platinum-carbon liquid used for the membranes that has to be mixed in a dust-free clean room. 97) Identify

(1) changes in the number of firms fi rms in an industry, (2) the relative investment by firms in product and process innovation, (3) the nature of competition  between firms, (4) the average average profit margin of  firms, and (5) the typical organizational structure. 3. Developing Fuel Cell Vehicles General Motors has built a fuel cell vehicle, which it plans to introduce commercially in 2011. (Fuel cell engines run on

the advantages and disadvantages that General Motors has at developing this technological innovation. Then take the perspective of an entrepreneur who has founded a new auto company; identify the advantages and disadvantages that the start-up has at developing the same innovation. Explain how the advantages and disadvantages of the entrepreneur and General Motors are the same and different.

NOTES 1. Adapted from Utterback, Utterback, J. 1995. Developing technologies: The Eastman Kodak story. story.  McKin  McKinsey sey Quarterly , 1: 131–144. 2. Narayanan, V. V. 2001. Managing Technology Technology and Innovation for Competitive Advantage. Upper Saddle River, NJ: Prentice Hall. 3. Dosi, G. 1988. Sources, procedures, procedures, and microeconomic effects of innovation. Journal of Economic Literature, 26: 1120–1 1120–1171. 171. 4. Markides, C., and P. P. Geroski. 2005. Fast Second: How Smart Companies Bypass Radical Innovation to Enter and Dominate New Markets . San Francisco: JosseyBass. 5. Mokyr Mokyr,, J. 1990. The Lever of Riches . New York: Oxford University Press. 6. Schil Schilling, ling, M. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill. 7. Dosi, Sources, procedures, procedures, and microeconomic microeconomic effects of innovation. 8. Cook Cooke, e, R. 2001. Dr Dr.. Folkman’s War: Angiogenesis and the Struggle to Defeat Cancer . New York: Random House. 9. Day Day,, D. 2004. Segway human transporter: More than a cool invention? Richard Ivey School of Business Case, Number 905M45. 10. 199 1998. 8. Marketing News, March 30. 11. Ettli Ettlie, e, J. 2000. Managing Technological Technological Innovation . New York: John Wiley Wiley.. 12. Miller, W. W. 2006. Innovation rules! Research Technology  Management, 49(2): 8–14; 1998. Marketing News, March 30. 13. Chandy Chandy,, R., and G. Tellis. 2000. The incumbent’s curse? Incumbency, Incumbency, size, and radical product innovation. Journal of Marketing, 64: 1–17. 14. Ibi Ibid. d. 15. Bower Bower,, J., and C. Christensen. 1996. Disruptive technologies: Catching the wave. Journal of Product Innovation Management, 13(1): 75–76.

16. Foster Foster,, R. 1986. Innovation: The Attacker’s Advantage. New York: Summit Books. 17. Ibi Ibid. d. 18. Brown, K., and A. Latour. Latour. 2004. AT&T will offer Internet phone calls in selected markets. Wall Street  Journal, March 30: B1, B2. 19. Chandy and Tellis, Tellis, The incumbent’s curse? 20. Clark, D. 2005. H-P team claims a milestone toward successor to the transistor transistor.. Wall Street Journal, February 1: B5. 21. Clark, D. 2007. H-P touts advance in chip technology. Wall Street Journal, January 16: B5. 22. Brown, R. 1992. Managing the “S” curves of innovation. Journal of Consumer Marketing, 9(1): 61–72. 23. Mitchell, W. W. 1989. Whether and when? Probability and timing of incumbents’ entry into emerging industrial subfields. Administrative Science Quarterly, 34(2): 208–230. 24. Rhoads, C. 2005. AT&T AT&T inventions fueled tech boom and its own fall. Wall Street Journal, February 2: A1–A12. 25. Zimmerman, R., R., and J. Bandler. 2004. Kodak sues Sony in patent dispute. Wall Street Journal, March 10: B4. 26. Allen Allen,, K. 2003. Bringing New Technology to Market . Upper Saddle River, NJ: Prentice Hall. 27. Clark, D. 2006. Freescale Freescale to sell memory chips using breakthrough technology. Wall Street Journal,  July 10: B8. B8 . 28. Belson, K. 2005. Cable’s new pitch: Reach out and touch someone. New York Times, May 8: 5. 29. Afuah Afuah,, A. 2003. Innovation Management. New York: Oxford University Press. 30. Conner Conner,, K. 1988. Strategies for product cannibalism. Strategic Management Journal, 9: 9–26. 31. Chandy Chandy,, R., and G. Tellis. 1998. Organizing for radical product innovation: The overlooked role of willingness to cannibalize. Journal of Marketing Research, 35: 474–487.

42

 

Technology Evolution

32. Chandy and Tellis, Tellis, The incumbent’s curse? 33. McGrath, R., I. MacMillan, and M. Tushman. Tushman. 1992. The role of executive team actions in shaping dominant designs: Towards Towards the strategic shaping of technological progress. Strategic Management Journal, 13:

57. Klepper Klepper,, S., and E. Graddy. Graddy. 1990. The evolution of  new industries and the determinants of market structure. Rand Journal of Economics, 21(1): 32. 58. Teece, D. 1986. Profiting from technological innovation: Implications for integration, collaboration,

137–161. 34. Foste Fosterr, Innovation: The Attacker’s Advantage. 35. Utter Utterback, back, J. 1994. Mastering the Dynamics of  Innovation. Cambridge: Harvard Business School Press. 36. Allen, Bringing New Technology Technology to Market. 37. Tam, P. P. 2004. As cameras go digital, a race to shape habits of consumers. Wall Street Journal, November 19: A1, A10. 38. Foste Fosterr, Innovat Innovation: ion: The Attacker’s Advantage Advantage. 39. Christensen, C. 1999. Innovation and the General  Manager. New York: McGraw-Hill. 40. Christensen, C. 1998. Valurec’s Valurec’s venture into metal metal injection molding. Harvard Business School Teaching Teaching Note, Number 698–002. 41. Day Day,, G., and a nd P. P. Schoemaker. 2000. Avoiding Avoiding the pitfalls of emerging technologies. In G. Day and P. Schoemaker (eds.), Wharton on Managing Emerging Technologies . New York: John Wiley. 42. Christ Christensen ensen,, Innovation and the General Manager. 43. Brown, K., and A. Latour. 2004. AT&T AT&T will offer Internet phone calls in selected markets. Wall Street  Journal, March 30: B1, B2. 44. Afuah Afuah,, Innovation Management. 45. Utter Utterback, back, Mastering the Dynamics of Innovation. 46. Afuah Afuah,, Innovation Management. 47. Ibi Ibid. d. 48. Teece, D. 1986. Profiting from technological innovation: Implications for integration, collaboration, licensing and public policy. policy. Research Policy, 15: 285–305. 49. Klepper Klepper,, S., and E. Graddy. 1990. The evolution of new industries and the determinants of market structure. RAND Journal of Economics, 21(1): 27–44. 50. Gort, M., and S. Klepper. Klepper. 1982. Time paths to the diffusion of product innovations. Economic Journal, 92(367): 630–653. 51. Freiberger Freiberger,, P., P., and M. Swaine. 1984. Fire in the Valley: The Making of the Personal Computer . Berkeley, CA: Osborne/McGraw-Hill. 52. Utter Utterback, back, Mastering the Dynamics of Innovation. 53. Mueller, D., and D. Tilton. Tilton. 1969. Research and development costs as a barrier to entry. The Canadian  Journal of Economics, 2(4): 570–579. 54. Utter Utterback, back, Mastering the Dynamics of Innovation. 55. Gort, M., and S. Klepper. Klepper. 1982. Time paths to the diffusion of product innovations. Economic Journal, 92(367): 630–653. 56. Ibi Ibid. d.

licensing and public policy policy.. Research Policy, 15: 285–305. 59. Adapted from Kirsch, D. 2000. Electric Vehicles and the Burden of History. New Brunswick, NJ: Rutgers University Press. 60. Damanpour Damanpour,, F., F., and S. Gopalakrishnan. 2001. The dynamics of adoption of product and process innovations in organizations. Journal of Management Studies, 38(1): 46–65. 61. Barras, R. 1986. Towards Towards a theory of innovation in services. Research Policy, 15: 161–173. 62. Damanpour and Gopalakrishnan, The dynamics of  adoption of product and process innovations in organizations. 63. Barras, Towards Towards a theory of innovation innovation in services. 64. Damanpour and Gopalakrishnan, The dynamics of  adoption of product and process innovations in organizations. 65. Barras, Towards Towards a theory of innovation innovation in services. 66. Damanpour and Gopalakrishnan, The dynamics of  adoption of product and process innovations in organizations. 67. Barras, Towards Towards a theory of innovation innovation in services. 68. Ibi Ibid. d. 69. Barras, R. 1990. Interactive innovation in financial and business services: The vanguard of the service revolution. Research Policy, 19: 215–237. 70. Afuah Afuah,, Innovation Management. 71. Schil Schilling, ling, Strategic Management of Technological Innovation. 72. Lynn, G., G., J. Morone, and A. Paulson. 1996. Marketing and discontinuous innovation: The probe and learn process. California Management Review, 38(3): 8–37. 73. Afuah Afuah,, Innovation Management. 74. Tushman, M., and W. Smith. 2002. Organizational technology. In J. Baum (ed.), The Blackwell Companion to Organizations. Cambridge, UK: Blackwell: 286–414. 75. Henderson, R., and K. Clark. 1990. Architectural innovation: The reconfiguration of existing product technologies and the failure of established firms.  Administrative Science Quarterly, 35: 9–30. 76. Ibi Ibid. d. 77. Ibi Ibid. d. 78. Christensen, C. 1997. The Innovator’s Dilemma. Boston: Harvard Business School Press. 79. Ibi Ibid. d. 80. Bower Bower,, J. 2005. Teradyne: The Aurora project.  Harvard Business School Case, Number 9–397–114.

43

 

Technology Evolution

81. Christ Christensen ensen,, Innovation and the General Manager. 82. Christ Christensen ensen,, The Innovator’s Dilemma. 83. Gilbert, C. 2003. The disruption opportunity. opportunity. Sloan  Management Review, Summer: 27–32. 84. Clark, Freescale to sell memory memory chips using break-

91. Christensen, C., M. Johnson, and D. Rigby Rigby.. 2002. How to identify and build disruptive new businesses. Sloan Management Review, Spring: 22–31. 92. Ibi Ibid. d. 93. Gilbert, The disruption opportunity. opportunity.

through technology. 85. Christ Christensen ensen,, The Innovator’s Dilemma. 86. Christensen, C. 1999. Hewlett Packard’s Merced Division. Harvard Business School Case, Number 9–699–011. 87. Christ Christensen ensen,, Innovation and the General Manager. 88. Christensen, C., F. F. Suarez, and J. Utterback. 1998. Strategies for survival in fast-changing industries.  Management Science, 44(12): S207–S220. 89. Ibi Ibid. d. 90. Paap, J., and R. Katz. Katz. 2004. Anticipating disruptive disruptive innovation. Research Technology Management, 47(5): 13–22.

94. Chandy and Tellis, Tellis, Organizing for radical product innovation: The overlooked role of willingness to cannibalize. 95. Christ Christensen ensen,, The Innovator’s Dilemma. 96. Bower Bower,, J. 2005. Teradyne: The Aurora project.  Harvard Business School Case, Number 9–397–114. 97. Boudette, N. 2006. GM hopes engine of of the future sells cars now. Wall Street Journal, November 29: B1, B2.

44

 

Technolog echnology y Adoption Adoption and and Diffusion Learning Objectives Technology Adoption and Diffusion:  A Vignette Introduction Distribution of Adopters Groups of Adopters S-Curves of Adopters Crossing the Chasm Identifying the Take-Off Stage  How to Cross Cross the Chasm Choosing the Customer Beachhead Strategy Market Dynamics Forecasting Demand

Information Diffusion Models  How Not to Do Do It The Bass Model Getting Down to Business: Diffusion of MP3 Players The Delphi Technique Product Diffusion Models The Importance of Complementary Technologies Substitution Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Identify

the different groups of adopters of new technology products and services, and understand the key factors influencing their willingness to adopt.

2. Explain

why adopters of new technology products and an d services tend to be normally distributed, and why the proportion of the market adopting a new technology product or service at a point in time is typically S-shaped.

3. Define

crossing the chasm , and explain why it is important for

firms. 4. Figure

out how to cross the chasm.

5. Explain

why forecasting demand is difficult, but important.

6. Define

technology diffusion, describe the typical diffusion pattern, and identify the factors that influence technology diffusion.

From Chapter 3 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved.

45

 

Technology Adoption and Diffusion

7. Understand

how information and product diffusion models predict the rate and functional form of diffusion.

8. Use

the Bass model to estimate the rate of diffusion of a new technology product or service.

9. Use

the Delphi Technique to estimate the rate of diffusion of a new technology product or service.

10. Explain

why complementary technology has a profound effect on technology diffusion.

11. Define

technology substitution, and explain how substitution affects the rate of adoption of new technology products and services.

12. Explain

why it is important to estimate how long it will take for technology substitution to occur, occur, and how technology substitution can affect the strategies of new entrants and incumbent firms.

Technology Adoption and Diffusion: Diff usion: A Vignette The concept of technology diffusion is well illustrated by the adoption of digital cameras. cameras. As Figure 1 shows, the sales of consumer digital digital cameras have grown dramatically since Apple Computer brought out the first one, the QuickTake 100, in 1994. The initial diffusion of the technology was slow from 1994 until 1996, when digital camera sales increased from zero to 400,000 units. The pace increased a little from 1996 to 1998, as sales grew to 1.1 million units per year. In 1999, the pace of diffusion accelerated, with sales almost doubling to 2.0 million units. Sales more than doubled in 2000, reaching 5.0 million units, and then almost doubled again over the next two years, to reach 9.3 million units by 2002. As a result, by 2003, digital camera sales in the United States had surpassed chemical film camera sales, with 12.5 million digital cameras being sold versus 12.1 million traditional film cameras. 1 By 2006, sales of digital cameras had grown to 30 million per year. Digital cameras also illustrate the concept of technology substitution. As sales of digital cameras have grown, sales of 35 millimeter traditional film cameras have fallen, dropping below 8 million in 2004, a decline of 20 percent in just two years.2 Even more dramatic has been the decline in sales of one of the most popular chemical film cameras, Kodak’s Advanced Photo System, which dropped from 10.5 million units annually in 2000 to less than 2 million in 2003.3 The substitution of digital camera technology for traditional camera technology also has led to a decline in the sales of photographic film. Sales of chemical film peaked at a high of 1.05 billion rolls in 2000 and have declined since. In 2005 alone, the major producer of photographic film in the United States, Kodak, faced a decline in traditional film sales of more than 25 percent.4

46

 

Technology Adoption and Diffusion

FIGURE 1 35

U.S. Digital Camera Sales

30    d    l   o    S   s    t    i   n    U    f   o   s   n   o    i    l    l    i    M

25 20 15 10 Digital Camera Sales

5 0         3         9         9         1

        4         9         9         1

       5         9         9         1

        6         9         9         1

       7         9         9         1

        8         9         9         1

        9         9         9         1

        0         0         0         2

        1         0         0         2

        2         0         0         2

        3         0         0         2

        4         0         0         2

       5         0         0         2

        6         0         0         2

Year

Digital camera shipments follow the typical accelerated pattern of technology diffusion over time. http://www.dataquest.com/press_gartner/ Source: Adapted from http://www.infotechtrends.com and http://www.dataquest.com/press_gartner/ quickstats/cameras.html.

I NTRODUCTION To formulate an effective technology strategy, you need to understand the patterns of  technology diffusion and customer adoption. You need to figure out the likely distri bution of adopters of your your products and services to target your your customer base effectively. You also need to follow particular strategies to transition from innovators to the majority of the market. Finally, you need to accurately forecast the growth of  demand for your products and services. This chapter discusses technology adoption and diffusion. The first section describes the distribution of adopters of new technology products and services, and explains how that distribution affects technology strategy. The second section explains how to transition from the initial adopters of your product to the majority of  the market. The third section discusses the dynamics of market growth, focusing on the factors that influence technology diffusion and substitution, and the tools that you can use to estimate the size of the market for, and the pace of diffusion of, your new products.

D ISTRIBUTION

OF  A DOPTERS

While you might be excited when your business sells a new product or service to its customers, the needs of justyou a few customers is usually not enough forfirst your companymeeting to succeed. Rather, probably need to achieve broad adoption (the decision of customers to purchase a new product or service) of your

47

 

Technology Adoption and Diffusion

new products or services by the mainstream of the market. If you don’t sell to the mainstream, you probably won’t be able to achieve sufficient sales volume to take advantage of economies of scale in production and distribution, making your cost structure uncompetitive. And you will likely get sidetracked into making custom products for your initial customers, will keep from dardized versions of your product. which Moreover, your you failure to developing transition tostanthe mainstream market will be particularly problematic if your business is new and is financed with venture capital; the cost of private capital virtually necessitates the reduction in costs and increase in margins that come from selling to the mainstream of the market. So, how can you achieve widespread adoption of your new product or service? Basically, you need to understand why and when different groups of customers adopt new products, and you need to make sure that your products possess the characteristics necessary to gain acceptance from each segment of the market. The first step in this process lies in understanding patterns of customer adoption of new technology products and services. These patterns are grounded in some s ome basic mathematics. Although there is some variation in the distribution of adopters of new technology products and services, 5 the single most common pattern is a normal distribution, as shown in Figure 2. Why? Because most patterns of human behavior are normally distributed, with a small portion of people doing things early, a small portion doing things late, and most people doing things in the middle.6 FIGURE 2

The Normal Distribution of  Adopters

   d   o    i   r   e    P   e   m    i    T   n    i   s   r   e    t   p   o    d    A    f   o   r   e    b   m   u    N

Innovators 2.5%

Two Standard Deviations Before the Mean

Early Adopters 13.5%

Early Majority 34%

One Standard Deviation Before the Mean

Mean Adopter

Late Majority 34%

Laggards 16%

One Standard Deviation After the Mean

The adopters of most new technology products are normally distributed; there are few innovators and early adopters, but also few laggards, with most people adopting in between. Source: Adapted from: Rogers, E. 1983. Diffusion of Innovations. Innovations. New York: Free Press.

48

 

Technology Adoption and Diffusion

For example, the distribution of adopters of CT scanners was normally distributed. Just less than 4 percent of hospitals saw the need for this technology right after it was introduced and became innovators. Another 14 percent were early adopters, purchasing on the heels of the innovators, while just over two-thirds were in either the early orafter late majority, and just under 7one-fifth were laggards who adopted the technology all of the other hospitals.

Groups of Adopters The different groups shown in the different sections of the adoption bell curve in Figure 2 adopt new technology products at different different times, and for different reasons. When a new technology product is first introduced, a small number of customers, called innovators, adopt it immediately. immediately. This segment of the market is very eager for new technology, technology, “needing” new devices as soon as they are made available. They are intrigued by technology and like to explore it. 8 While most potential customers require a lot of information about the potential value of a new product or service before adopting it, this portion of the market is savvy enough to evaluate new technology products and services with little information, using intuition to make decisions about a new product’s value. They often pursue products before their formal marketing, sometimes even buying prototypes and  beta versions. Take DVD players as an example. You probably probably know someone who  bought one when they were first introduced because that person just had to have the technology to play with. Innovators are not very price sensitive, and are willing to pay a lot of money for new technology products and services. 9 For instance, some people in California have adopted electric cars even though these cars cost much more to lease than cars with internal combustion engines, and there are few places to recharge the vehicle  batteries. 10 A second group of customers, called the early adopters, follow the innovators. This group of customers is larger than the initial group of innovators. Why? Because more people are willing to purchase new products and services as information about them becomes known, and uncertainty about their t heir value declines. And the adoption of  new technology products and services by the innovators generates information about the value of the new products to potential early adopters, which helps them to make the adoption decision. The early adopter segment of the market does not have as much m uch knowledge of the potential value of new products as the innovator segment, but it is technically savvy, savvy, and responds to marketing with a great deal of technical content. The early adopters are an important group because the effort to sell to them often provides evidence that can be used to sell products to the majority of the market. Take, for example, the experience of a medical device company named Imalux that has developed a new technology called optical coherence tomography, tomography, which can be used to capture high-resolution images of tissue that cannot be seen with alternative technologies. Imalux is targeting urologists because they already use fiber optic devices to see inside the body. body. The company hopes to use information from the opinion leaders in that segment to generate evidence of effectiveness, which can be used to formulate the value proposition for the majority of the market.11 A third segment of of the market is is the early majority, which tends to adopt a new technology product slightly before the average for the market. This segment sees 12

value in technology but isreference driven by practical considerations. It views the early adopters as an important group and looks at their adoption decisions as 13 indicators of the value of the new technology. For instance, the people who waited

49

 

Technology Adoption and Diffusion

until cell phones had shrunk to handheld sizes and were reasonably priced were the early majority of the cell phone market. To convince the early majority to adopt your product, you need to change your product’s features and the way that you sell it. The early majority wants to 14

see a whole product solution and a value proposition before buying. So youmeswill probably need to add features to your product and revamp your marketing sage to sell to this group. Your new advertising message should not emphasize technical content, but should focus on the completeness and ease of use of your product. You also should try to achieve a dominant position in the market. The early majority prefers to purchase from the market leader because the leader’s products tend to be the most reliable, and the most likely to provide support in the aftermarket. 15 A fourth segment segment is is the late majority, and is roughly the same size as the early majority. This segment is often more skeptical about the value of technology than the early majority, and tends to adopt it only after strong evidence shows that the value of adoption is greater than the cost of adoption, and that other customers have h ave 16 adopted the technology successfully. In general, adoption by the late majority is driven by considerations other than the value of the technology itself. 17 For instance, the late majority adopted CDs in place of cassettes and records when it  becam e diffi  became difficult cult to buy musi musicc in the older form formss and the avai availabi lability lity of musi musicc  became a factor in the type of devices used to play music. The late majority is price sensitive and has strong needs for service and support. To sell to this segment successfully, not only must you offer a whole product,  but also your whole product must not be expensive, and must be reliable and easy ea sy to use. The most important advertising message for this segment is reliability and low cost.18 A final market segment, called called the laggards, follows the late majority. This segment will not adopt a new technology product until it is well established, preferring preferring to avoid adoption of these products as long as possible, and often adopting only when the alternatives to adoption are no longer available. 19 This group is smaller than the majority of the market, and is generally equal in size to the innovators and early adopters combined.20 I bet you know a laggard in some technology area. Perhaps your parents have never caught on to CDs and MP3 players and still have cassettes in their basement. In the business of music-playing devices, those people still playing cassettes on their stereos would be considered laggards.

S-Curves of Adopters Understanding that the typical pattern of adoption of new products and services is normally distributed is important for you, as a technology strategist, for several reasons. First, it points out that differe different nt groups of customers adopt new products at different points in time for different reasons. Innovators typically purchase new products because they need to explore the uses of new technology, while laggards usually adopt them because they have no alternative. The motivations of the different market segments indicate that you need to design variations of your new product to meet the needs of these different segments.21 Second, this distribution provides important information about the right promotional for, and yourfor new For theand innovator may not bestrategy necessary, necessary theproduct. early adopters the earlysegment, majority,advertising majority, promotion  based on referenc references es from satis satisfied fied custo customers mers might be very effecti effective. ve. However,

50

 

Technology Adoption and Diffusion

for the late majority of the market, you need to shift to more informational and price-based advertising.22 Third, the normal distribution of adopters indicates the appropriate pricing strategy for your new product. For the innovators and early adopters, who are not very price-sensitive, a high price might betothe way to go. For the majority of the ket, you might want to reduce your price spur adoption. For the laggards, youmarwill need to cut your price because this group is quite price-sensitive. Take, for example, the change in online shopping between the late 1990s and today. In the late 1990s, Internet shoppers tended to be young, financially well-off  males who were technology savvy and price insensitive. They were quite willing to purchase products from independent online retailers like Amazon.com. Today, as many more people make online purchases, the typical customer is female, less welloff, and less technology savvy. For these customers, lower prices and brand name reputations attract them, which has allowed companies like the Gap and Wal-Mart to catch up to the online retail pioneers.23 Fourth, understanding the distribution of adopters provides you with information that will help you to estimate demand growth growth over time. As As Figure 3 indicates, the normal distribution of adopters translates into an S-shaped pattern of demand growth over time because the bell curve shows the rate of adoption while the S-shaped curve shows the cumulative adoption.24 (For those of you used to thinking in calculus terms, the bell curve is the derivative of the S-curve, and the S-curve is the integral of the bell curve.) FIGURE 3

The Adoption S-Curve

Slowing of Market Growth 80

60

  g   n    i    t   p   o    d    A    t   n   e   c   r 40   e    P

Acceleration of Market Growth

Low Rate of Market Growth

20

0

5

10

15

20

25

30

35

Time

The percent of the market that has adopted a new technology product at a given point in time tends to follow an S-shaped pattern, displaying slow growth initially, then accelerating, and then decelerating as the market becomes saturated.

51

 

Technology Adoption and Diffusion

The S-shaped pattern indicates that there is an acceleration point at which a market takes off. Consequently, you need to be prepared for this acceleration by ensuring that you have adequate supply of raw materials and labor, and sufficient manufacturing and distribution infrastructure, to meet demand. Moreover, even when adoption patterns do not follow a normal distribution, use deviations from that baseline to forecast the growth in demand for your you newcan technology product over time. Fifth, the normal distribution of adopters tells you something about the financial attractiveness of a market at different points in time. In the first stage of the product life cycle, when customers are innovators and early adopters, revenues are low and costs are high, making profit margins relatively small. In the second stage of the life cycle, when you transition to serving s erving the majority of the market, revenues rise faster than costs, and profit margins increase.25 In the third stage of the life cycle, when customers are laggards, revenues fall and costs increase, leading profit margins to decline.26  Key Points

• The most common pattern of adoption of new products and services is a normal • distribution. Adopters can be divided into five segments: innovators, early adopters, early majority, late majority, and laggards; each group adopts products at different points in time, and for different reasons. reasons. • The normal distribution of adopters leads to an S-shaped pattern of cumulative adoption over time. • Markets for new technology products “take off” when they transition from early adopters to the early majority of the market. • The normal distribution of adopters is important to technology strategy because it highlights the different motivations of each group of adopters; illustrates the right promotional strategy to reach each of them; provides information about how to price products; helps to estimate demand growth; and provides information about the financial attractiveness of the market over time. • The adoption S-shaped curve is different from the technology S-curve. The adoption S-curve measures the percent of the market adopting a new product as a function of time. The technology technology S-curve measures the level of of performance of a technology as a function of the amount of effort put into its development.

C ROSSING

THE   C HASM

The discussion of the distribution of adopters in the previous section indicates the importance of transitioning to the majority of the market, or what many people call the “mass market.” For most products, markets are too small for you to be successful  by just selling to innovators and an d early adopters. To achieve an adequate return on your investment in product development, you need to sell to the mass market. If you don’t, you will end up engaged in costly customization for the innovator and early adopter Moreovertake Moreover, , the advantage faster you move to the mass market, sooner you can sell segments. in larger volume, of scale economies, andthe make greater profits. So how do you transition to the mass market?

52

 

Technology Adoption and Diffusion

Identifying the Take-Off Stage The first step in reaching the mass market is identifying that a new product has reached the take-off stage. You You can figure this out in a couple of differe different nt ways. First, you can look at the past demand for the product. Is the rate of change in demand accelerating? If it is, that signals that a product is reaching the take-off stage. Second, what do new customers for the product look like and how do they differ from existing customers? When the market is about to take off, the customer base for a product begins to shift away from the innovators and early adopters. If new customers are focused on the value proposition from purchasing your product or service or care a lot about the ancillary attributes of your product package, then your product or service is probably reaching the take-off stage. You don’t have to be completely reactive about your product reaching the take-off  stage; there are things you can do to push your product to take off. First, any actions that you take—developing a marketing message based on the value proposition, getting early adopter testimonials, etc.—that make the mainstream market more interested in the product will increase the odds that the product will take off. Second, changes that you make to improve your product and lower its cost will make the product more attractive to a wider range of the market, and more likely to take off.

How to Cross the Chasm Marketing consultant Geoffrey Moore termed this transition crossing the chasm (see Figure 4) because of the differences in the adoption decisions of early adopters and the majority of the market. 27 As the previous section of the chapter indicated, to sell your new product to the majority of the market, you need to show how it provides value to customers, which you might not need to do with the innovators or early adopters. For instance, selling your CAD/CAM software to the majority of the market could require you to show how your customers will speed up their design process and save money by adopting your product, rather than just demonstrating how cool the software is. Moreover, you need to develop a complete solution to your customers’ problems—things like manuals, customer service, set-up assistance, and so on—  because the majority of the market seeks solutions to its problems, rather than just pieces of technology. If you don’t provide a complete solution, you are likely to fail to cross the chasm. For instance, Nokia experienced problems crossing the chasm when it introduced the N800 Internet Tablet, a small computer. The company was relying on open source software developers to provide a calendar application, a way to synch data to a desktop computer, and the software necessary to record pictures, so the company didn’t develop these features. But mainstream customers were expecting Nokia to provide a complete product with all of these features already present, and so didn’t adopt the product.28 Furthermore, you will need to pursue a vertical marketing strategy (focusing on customers in a single industry) rather than a horizontal marketing strategy (serving customers in multiple industries at the same time). The majority of the market seeks references from adopters they know, and this is not likely if those customers are in different industries. It is also less expensive and easier to create these references if  you focus on a single industry at a time because of the links that exist between cus29

tomers in the same Lastly, if you adopt horizontal marketing strategy, your resources will industry. be spread too thinly for you to adevelop a complete solution to 30 your customers’ problems and demonstrate the value of that solution to them.

53

 

Technology Adoption and Diffusion

FIGURE 4

Crossing the Chasm

   d   o    i   r   e    P   e   m    i    T   n    i   s   r   e    t   p   o    d    A    f   o   r   e    b   m   u    N

The Chasm

Innovators

Early Adopters

Early Majority

Late Majority

Laggards

Time

There is a chasm between early adopters and the early majority that results from differences in the demands of marketing to the two groups of customers. Source: Adapted from Moore, G. 1991. Crossing the Chasm. Chasm. New York: Harper Collins.

Choosing the Customer Because you need to concentrate on a single vertical market to transition to the early majority, majority, you need to figure out which vertical market to focus on first. You should target the one that has the greatest need for your new product because your ability to demonstrate the value of that product will be greatest in this market. 31 For example, Dell first chose to sell computers to business customers  because they th ey had a greater great er need than consumer consumerss for low-priced low-pr iced high-end high- end computcomput ers with little tech support. A vertical market has a great need for your new product if the new product product would improve your customers’ productivity, cut their costs, or allow them to do something that they otherwise could not do. 32 For example, UPS has a great need for a computerized system to track the location of its vehicles because such a system would allow the company to identify the most precise routes to destinations, thus cutting the cost of delivering packages. To determine which vertical market has the greatest need for your new product, you need to estimate its value to different markets. The best way to do this is to estimate the time it takes to pay back the cost of the product in each market. Take, for example, a product that makes roof shingles reflect light. The benefit of  this product can be estimated because reflecting light reduces the cost of cooling a  building.  buildin g. Therefore Therefore,, value of the product can be measu measured red by the paybac payback k from reduced energy costs. Given the cost of the shingles, the average payback time is thr three ee ye years ars.. 33 But this is the average. You can examine different vertical markets—new home builders, commercial builders, and roofing companies—to see which one has the shortest payback time.

54

 

Technology Adoption and Diffusion

If you cannot estimate the payback from purchasing the product, you might  be able to quant quantify ify the value to the differe different nt verti vertical cal mark markets ets by esti estimati mating ng the amount of savings that customers achieve from using the product. For instance, the developers of online sales systems were able to figure out which type of retailers— clothing, books, insurance, airline tickets, and the so on—had greatest for their technology because they could measure labor costthe savings in need different vertical markets. Unfortunately, estimating the value of a product to different vertical markets is not always this straightforward. Sometimes, the benefits of the new product cannot  be captured in numbers. For example, the value of a new service that helps companies to improve labor relations may be hard to quantify because it lies in having “happier” employees. In general, when it is difficult to quantify the value of a new product to potential vertical markets, it is also difficult to identify the vertical market with the greatest need to adopt that product.

Beachhead Strategy For new companies to successfully cross the chasm, it is sometimes necessary for them to segment the early majority of the market, looking for a niche whose needs are not well met. By focusing on this segment of the market, the company can make headway.34 An example of this process at work is the effort to introduce VOIP telephone service. Most customers were reluctant to adopt this new technology initially because it did not provide many of the features of traditional telephone service. However, However, some segments of the market were not well served by the traditional telephone companies and saw VOIP as a solution to their problems. problems.35 These firms became the initial customers of new companies that provided provided VOIP telephone service. service. Once the new companies established this beachhead in the market, they were able to expand to other market segments by improving the technology underlying their products and making it useful to a wider range of the market.36  Key Points

• Because the early majority adopts products for differe different nt reasons from early

•

• •

•

•

adopters, companies often find it difficult to transition to the mainstream of the market. Providing a complete solution, demonstrating the value proposition, and focusing on a single vertical market are all things companies need to do to transition to the early majority. Firms are most successful at transitioning to the early majority if they focus on the vertical market with the greatest need for their product. The vertical market with the greatest need for a new product is the one in which the introduction of the product would lead to the largest increase in customers’ productivity,, the largest decline in customers’ costs, or the greatest opportunity productivity for customers to do something that they otherwise could not do. To determine which vertical market has the greatest need for a new product, you can estimate the time it takes customers to pay back the cost of the product or the amount of savings the customer gets from using it; however, however, the value of  a new product to customers cannot always be quantified. To successfully cross the chasm, it is sometimes necessary for new companies to segment the early majority of the market and focus on the portion that is underserved by existing products.

55

 

Technology Adoption and Diffusion

M ARKET   DYNAMICS Although it is very difficult to accurately forecast anything that happens in the future, you still need to estimate what demand will be for your new products for several reasons: • First, demand forecasts help you to determine how much to produce, allowing you to avoid underproduction and the loss of potential customers to your competitors, as well as overproduction, and the corresponding need to cut your price and profits. • Second, demand forecasts help you to project your future costs in businesses  based on economies economies of scale. In In businesses with high fixed costs, unit costs decrease as the volume of production increases, making your costs a function of  the size and rate of sales growth.37 • Third, demand forecasts help you to determine the payback on your investment in product development. Because the costs of developing a new product tend to  be fixed, the payback on investments investments in product development development depends depends a great deal on amount and timing of sales. • Fourth, demand forecasts help you to make pricing and advertising decisions38  because they provide provideoninformation informati sensitivity, and the effect of advertising sales.39 on about customers’ price sensitivity, • Fifth, demand forecasts help you to determine the competitiveness of your market.40 Stagnating and shrinking markets are more competitive than growing markets because sales in the former have to come at the expense of competitors,  but sales in the latter latter can come from from new customers. customers.

Forecasting Demand As Figure 5 illustrates, predicting predicting anything that will happen in the future isn’t easy. easy. This holds true for market demand as well as anything else. Because markets are dynamic, you can’t estimate future demand solely on the basis of the current market size. In fact, for many new technology products and services, a huge market tomorrow may not even exist today. Take, for example, the market for Internet search, which is enormous, but did not even exist in the early 1990s. Moreover, forecasting demand depends a lot on the timing of customer adoption. To understand why, just think about the difference in two forecasts of a billion dollar FIGURE 5

Predicting the Future

While it’s easy to predict past success, it isn’t easy to predict future winners. Source: The Piranna Club, August 19, 2006.

56

 

Technology Adoption and Diffusion

market—one which indicates that this market will emerge in 5 years, and the other, which indicates that it will emerge in 150 years. The market size at different points in time over the next 150 years will be completely different in the two forecasts. Furthermore, the rate of growth of demand over time is not linear. (Remember the S-shaped curve of adoption from theshift firstfrom section of the chapter?) Because demand growth accelerates when markets the early adopters to the early majority, a key to forecasting demand accurately lies in identifying when this point of acceleration will occur. Finally,, forecasting demand depends a lot on the accuracy of information about Finally the factors that influence diffusion patterns. ( Diffusion is the rate at which a new technology product becomes adopted by potential users. 41) As we will see later in this chapter, both the functional form and the rate of diffusion can be affected by a variety of factors, including the external environment, the proportion of innovators and imitators in a market, and the nature of the product.

Information Diffusion Models As the previous section indicated, you need to know something about the factors that affect the diffusion of your new products to forecast demand accurately. Information diffusion models explain that the functional form of diffusion is primarily a result of the distribution of innovators (customers who learn about new products from sources other than previous adopters, such as from advertising) 42 and imitators (customers who learn about new products from previous adopters).43 In all markets, some s ome people are innovators and adopt new products independently of learning information from others, while most people are imitators, and adopt new products when they learn that others have adopted them. 44 The distri bution of innovators in novators and imitator imitatorss varies va ries across market marketss and a nd affects the shape of  the diffusion curve. As was explained in the beginning of the chapter, when there are few innovators and many imitators, diffusion follows an S-shaped pattern. 45 In the beginning, only small numbers of innovators purchase new technology products and services, and the rate of diffusion is slow. However, as imitators learn about the new product from the innovators, the rate of diffusion accelerates. As the market becomes saturated, the rate of diffusion slows, and ultimately reaches an asymptote. 46 (While you might start to think that the only shape that professors of technology strategy can draw is an “S,” the diffusion S-curve is different from the technology S-curve and the adoption S-curve described earlier in this chapter. The diffusion S-curve measures the number of adopters as a function of time. It is  based  base d on the idea that some part partss of the mar market ket adop adoptt inde independe pendently ntly from the decisions of other users, while other parts of the market do not. The adoption Scurve measures the percent of the market adopting a new product as a function of  time. It is based on the idea that the shape of cumulative adoption is related to the shape of the distribution of adopters, with a normal distribution of adopters yielding an S-shaped curve of cumulative adoption. The technology S-curve measures the level of performance of a technology as a function of the amount of effort put into its development. It is based on the concept that performance improvements require a lot of effort initially and later run into diminishing returns.) The diffusion of new technology products and services is not always S-shaped. For instance, diffusion patterns look more like a convex curve than an “S” if there are many innovators and few imitators. This happens when potential adopters are not

57

 

Technology Adoption and Diffusion

FIGURE 6

The Diffusion of  the CT Scanner and Cable Television

25   s   r   e   s    U    l   a    i   n    F    f   o   e   g   a    t   n   e   c   r   e    P

20 15 10 5 0 0

5

10

15

20

25

15

20

25

Year

25   s   r   e   s    U    l   a   n    i    F    f   o   e   g   a    t   n   e   c   r   e    P

20 15 10 5 0 0

5

10 Year

The diffusion of the CT scanner took the standard S-shape, S-shape, whereas the diffusion of cable television was more of a convex curve. Source: http://andorraweb.com/bass/index.=?show[examples]=1.

closely connected, leading the decisions of earlier adopters to have little influence on the decisions of later ones. 47 Figure 6 provides the example of the diffusion of the CT CT scanner,, which followed an S-shaped pattern, and cable television, which followed a scanner pattern best described as a convex curve.

How Not to Do It The previous section indicates that you need some method for estimating the size of  the market for your new products. Unfortunately, the most common way that entrepreneurs and managers do this is not very accurate. The typical approach involves looking up the size of a similar market and using that number as the estimate of the size of the target market. For instance, take a company that has developed a new communications device called e-mail. At the time that the company has developed the first e-mail communications system, no one has a way to send messages from computer to computer. So the manager responsible for analyzing the market size estimates it by looking up how many telephone calls people make each day, and uses that information to estimate the number of e-mail messages that people will send.

58

 

Technology Adoption and Diffusion

However, to use information on the telephone market to estimate the market for e-mail, you have to assume that the purpose of e-mail is to replace the telephone, making e-mail a substitute for that product, rather than a substitute for regular mail, faxes, face-to-face conversations, or other forms of communication. If the true purpose of email to substitute for other products like face-to-face conversations, mail, or faxes,isthen market size estimates for e-mail that are based solely onregular the size of the market for telephone calls are not going to be very accurate. Moreover, e-mail is often a complement to telephone communication because people often send e-mail messages to clarify telephone calls, and vice versa. As a result, by assuming that e-mail substitutes for the telephone in the evaluation of market size, you will likely get a number that has nothing to do with the size of the actual market for e-mail communication.

The Bass Model So what should you do to estimate demand if just looking at an analogous market won’t work? You You should use quantitative or qualitative methods for forecasting the diffusion of new technology products. One quantitative tool for forecasting the diffusion of new technology products that many companies use is the Bass diffusion model.48 In this model, diffusion diffusion patterns are a function function of the size of the market, the rate of adoption by innovators and imitators, and the proportion of adopters in the previous time period.49 They can be measured by the following formulas:50 The likelihood of purchase by a new adopter at time period t is p (q/m)n t 1,

where p  is the likelihood that an innovator will adopt, q  is the likelihood that an imitator will adopt, m is the total number of adopters that will never be exceeded, n t 1 is the cumulative number of adopters of the product through the previous time period.

The number of new adopters during time period t, S(t), equals [p (q/m)n t 1][(m-n t 1)]

That is, in the Bass model, the number of adopters in a period is measured by the sum of the probability that an innovator adopts the new product plus the probability that an imitator adopts, divided by the size of the market, and multiplied by the proportion of the market that has already adopted the product, all multiplied  by the size of the market, mar ket, less les s the amount of the market mar ket that tha t has already adopted. ad opted. For instance, suppose the rate of adoption by innovators is 0.10 and the rate of  adoption by imitators is 0.50, the market size is 10 million, and no one has yet adopted the product. The number of adopters in the year would be predicted to be: (0.10  (0.5/10)  (0))  (10  0)  0.1  10  1.0 million customers. So where do you get the numbers to plug into the Bass model? Typically Typically,, users estimate the market size by conducting a survey of potential customers or by examining secondary data. And they estimate the likelihood that innovators and imitators will adopt by extrapolating from historical data on the diffusion patterns of similar products.51 For example, you could create estimates of the rate of innovator and imitator adoption of HDTV by extrapolating from published data on the rates for those two 52

groups taken from studies of color televisions, VCRs, and other consumer electronics. The basic Bass diffusion model can be modified to include a variety of factors that affect the diffusion of new technology products, such as price,53 and changes in the

59

 

Technology Adoption and Diffusion

size of the potential market.54 You can estimate the diffusion of complementary products (like washing machines and dryers) by using a variant of the model that makes the diffusion of one product contingent on the diffusion of another. 55 Similarly, you can capture the effect of substitution by future generations of products, and accom56

modate repeat purchases with other variants. You can even estimate market share  by linking linking the the Bass model to discrete discrete choice models that compare compare the likelihoo likelihood d that a customer will choose between your product and your competitors’ products, using information from a conjoint analysis on different product attributes.57 How effective are these models at predicting the adoption of new technology products and services? They are pretty good, but not outstanding. And they are better for some products products than for others. The Bass model is most accurate at predicting the diffusion of consumer durables (consumer products, like appliances, whose value is not exhausted in a year), than other types of products.58 Moreover,, the Bass model has important limitations, which you need to consider Moreover if you are going to use it. 1. You cannot use the Bass model to estimate diffusion at the very beginning of a product’s life because the model uses information from previous periods to estimate diffusion in subsequent periods. 2. The accuracyofofsize the of Basthe Bass s model’s predictions on estimate the accuracy of your assessments potential market. If depends you cannot the size of  the potential market, or if your estimate is wildly inaccurate, then the Bass model isn’t going to be much help to you. 3. The Bass model assumes that the diffusion of a technology product depends on only demand-side factors.59 If the size of a market, or the rate of adoption of a product, are determined by supply-side factors, such as the availability of raw materials or production capacity, capacity, then the results of a Bass model analysis will  be very inaccurate. inaccurate. 4. The accuracy of the Bass model is much lower when competing technologies are  being intro introduced, duced, or when when economic economic shocks are occurri occurring, ng, because because the model model has has no mechanism for estimating the effect of competitive factors on new product adoption, or ways to incorporate the effects of external economic shocks. 5. The accuracy of the Bass model declines as you get further away in time from the initial adoption point. GETTING DOWN TO BUSINESS

Diffusion of MP3 Players 60

Let’s look at an example of how a company might use the Bass Diffusion Model. Suppose you are an entrepreneur who has founded a company to make MP3 players at the start of that industry. You need to create estimates of market size and growth to select a revenue model and to justify your financing strategy to potential investors. So you decide to use the Bass model to create those estimates. To calculate the diffusion of your product, you

suppose that you conducted a survey of potential customers and found out that this number is 40 million people. You then need to estimate the rate of adoption  by innovators and by imitators. You can create these estimates by looking at analogous products to the MP3 player—things like the portable CD radio and the cellular telephone. If you find the average rate of  adoption for innovators and imitators of these kinds of  products is 0.01 and 0.40, respectively, you can gener-

first need to estimate the total size of the market. Let’s

ate the diffusion pattern shown in Figure 7.

60

 

Technology Adoption and Diffusion

FIGURE 7

A Demand Forecast for MP3 Players

Year 1 2

S(t) 0.400000 0.554400

N(t) 0.400000 0.954400

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

0.763107 1.040330 1.399500 1.848527 2.381583 2.967611 3.539099 3.990362 4.198712 4.074105 3.616152 2.931585 2.184659 1.516297 0.996252

1.717507 2.757837 4.157336 6.005863 8.387446 11.35506 14.89416 18.88452 23.08323 27.15733 30.77349 33.70507 35.88973 37.40600 38.40225

Adoption Rate

45 40    t   e 35    k   r   a 30    M   e 25    h    t    f   o 20    t   n   e 15   c   r   e    P 10

5 0 0

1

2

3

4

5 6

7

8 9 10 11 12 13 14 15 16 17 Time

The Bass model allows you to forecast demand for new products.

The Delphi Technique Because of these limitations of the Bass model, some experts believe that you are better off making your organization flexible so that it can respond to changes in market demand, rather than trying to predict how demand will change over time. However, you do need to have some idea of where demand is going to figure out how much to produce when first introducing a new technology product. For this reason, some experts suggest that you use more qualitative methods to forecast demand. One important qualitative method that many companies use is called the Delphi Technique. Technique. As Table Table 1 shows, with the Delphi Technique, Technique, experts TABLE 1

The Process for Using the Delphi Technique The use of the Delphi Technique involves the simple 10-step process, outlined here.

STEP

ACTION

1 2 3 4 5 6 7 8 9 10

Pick a facilitator Identify an expert group Create an initial list of criteria Have the experts rank criteria Estimate the mean and standard deviation Have panel re-rank the newly ordered criteria Identify pr preferences and co constraints Have th the pa panel rank al alternativ ivees by by co constraints an and pr preferen encces Analyze the results and return them to the panel Rep epea eatt th the ran ranki kin ng pro proce cess ss unt ntil il yo you u ac ach hie iev ve sta stab bil ilit ity y on th thee ran ranki kin ngs

Source: Adapted from: Cline, A. Prioritization Process Using Delphi Technique, Technique, downloaded from http://www.carolla.com/wp-delph.html.

61

 

Technology Adoption and Diffusion

are selected and asked anonymously for their estimates of the likelihood of particular outcomes occurring. The participants return their estimates to a coordinator coordinator,, who compiles them. The summary data outlining the mean and range of viewpoints is then returned to the respondents who are asked for new estimates in light of the 61

information presented by the The Delphi Technique Technique hasother beenexperts. used with success in a variety of settings, and it is an important tool for you to use to identify potential technological trends that might impact the development of your new products and services. However, you need to be aware of several major weaknesses of the technique. First, the method is sensitive to the precision of the questions that you ask. If you ask imprecise questions, the respondents might not reach a consensus, and, if they do, that consensus might not be about the major trends that you are seeking to understand. Second, the Delphi Technique Technique is sensitive to variance in the expertise of the respondents. If the respondents are not experts on the topic about which they are being queried, then the technique provides little more than a consensus of the ignorant, which is of little use to you as a technology strategist. 62 Third, the validity of the technique is limited by the intervention of unexpected events that the experts do not incorporate into their analyses. Because the Delphi Technique is built on consensus, it is only accurate at predicting things that most people expect to occur.

Product Diffusion Models Some researchers propose product diffusion models that are different from the information diffusion models, like the Bass model, described previously previously.. Product diffusion models look at product characteristics to explain their rate of diffusion and have found a variety of different characteristics to be important. First, the greater the benefit that a new product provides to customers and the lower its cost, the faster it will diffuse diffuse because customers customers have more motivation motivation to 63 adopt a product product that benefits benefits them more and and costs them less. For instance, asymmetric DSL DSL technol technology ogy diffused diffused slowly because because the cost to deploy deploy the technology technol ogy was high in comparison to alternative alternative technologies technologies for transmitting digital data.64 The cost of a new technology product is affected by the degree of change that customers have to make to adopt it, leading diffusion to slow as the needed amount of change increases. For example, 3G mobile phones diffused to wireless carriers more slowly than previous generations of mobile phone technology because the adoption of 3G required the carriers to purchase a new wireless spectrum, and to create new signal stations, whereas previous mobile phone technologies did not require such significant upgrades.65 The effect of the degree of change on the rate of diffusion is not limited to industrial products. The diffusion of CDs was slower s lower than the companies producing them had expected because customers saw replacing or copying their cassette tapes as a cost to adopting the new technology. 66 Sometimes, companies need to produce several generations of new products  before  befo re tho those se prod product uctss offer enou enough gh ben benefit efit to cus custom tomers ers to spu spurr diff diffusio usion. n. For example, digital cameras did not begin to diffuse quickly until after manufacturers had produced several generations of them, and their level of resolution had reached that of chemical film cameras. Second, the amount of learning necessary to understand the benefits of a product will slow its rate of diffusion because people must understand the value

62

 

Technology Adoption and Diffusion

of new products and services to adopt them. 67 Thus, the provision of information about a new product (including advertising) and the opportunity to test it enhance the rate of diffusion.68 Third, the perceived risk of a new product lowers its rate of diffusion. 69 If new 70

products are perceived as risky because competing technical standards exist or  because the products prod ucts are novel, potential adopters often delay d elay adoption a doption to gather ga ther more information.71 For example, diffusion of high-definition DVDs has been slow  because manufacturers have failed to t o agree on a common technical standard, and consumers have been unwilling to risk selecting a format that is not the industry standard.72 Fourth, the characteristics of the adopters affect the rate of diffusion of new products. Products diffuse faster when adopters are tolerant of uncertainty because this attribute makes them more open to new product ideas. They also diffuse faster when adopters are wealthier because wealth provides a cushion against poor adoption decisions.73 Moreover, new products diffuse more quickly when adopters are geographically proximate, or tightly connected through social networks because these attributes accelerate information transfer trans fer..74 For instance, pharmaceutical firms, like Merck, try to get their new drugs into the hands of well-connected doctors because the interconnectedness of the medical community makes diffusion of medically related products very rapid. Fifth, aspects of the environment in which the adoption decision is being made affect the rate of diffusion of new products.75 For instance, new products that involve capital investment diffuse faster as interest rates decline because the cost of  adoption decreases as interest rates go down. Moreover, this effect is larger for more expensive products because the use of financing increases with the cost of products. Social factors also affect the diffusion of new products. When new products are socially acceptable, they diffuse more quickly than when they face social opposition. For example, a product that allows people to use a Web cam and Internet technology to record their prayers and send them over the Internet to a religious site has diffused slowly because it has faced opposition from religious leaders who think that the use of high technology for prayer demeans religion.76 Political factors, too, influence the rate of diffusion of new products. For instance, when an important political group opposes the adoption of a new product, its diffusion is slowed. radio, for instance, diffused slowly the AM got the FederalFM Communications Commission (FCC)because to earmark the radio 42 to industry 50 MHz radio frequency range to television, to require a new frequency for FM radio, and to allow programs to be broadcast on both AM and FM radio.77 Political opposition can even stop the diffusion of a new product. Consider what happened to the diffusion of nuclear power in the United States after the Three Mile Island disaster led to a dramatic rise in political action against the nuclear power industry industry..78 Because people became afraid of a nuclear disaster and public support for nuclear power dropped dramatically, the diffusion of  nuclear power stopped.79

The Importance of Complementary Technologies The nature of an innovation affects its rate of diffusion. In general, new products based on discrete technologies diffuse faster than ones based on systemic technologies  because new produ  because products cts based based on systemic systemic technologi technologies es can only only diffuse diffuse as fast as their their slowest diffusing component. For example, the diffusion of music downloads was slow until MP3 players were developed developed and the downloaded music could be made portable.

63

 

Technology Adoption and Diffusion

For systemic technologies, an important part of managing a strategy of technology diffusion involves coming up with ways to ensure that complementary technologies will develop, and figuring out when those technologies will come to fruition. For example, General Motors and Ford have prototype fuel cell vehicles that they arethat planning to introduce tovehicles the market in 2011. However, companies recognize the adoption of these depends a great deal onboth the ability to refuel them, making the complementary technology of a system of hydrogen refueling stations of central importance. For this reason, Ford and General Motors, among other automakers, are working with the major oil companies to establish a network of hydrogen refueling stations throughout the country so that they can introduce their new vehicles in the next five years.80

Substitution As was described in the opening vignette to this chapter, the diffusion diffusion of new products and services is often affected by substitution. ( Substitution is the replacement of  one technology by another that can achieve the same objective. 81 ) While some observers argue that all new products and services are substitutes for some existing product or service, substitution is more direct in some cases than in others. For example, fiber optic cable is a very clear substitute for coaxial cable because telecommunications firms now use the former for almost all purposes that they previously used the latter. Similarly, DVD players have substituted for laser disk players, and effectively stopped their diffusion.82 Substitution can have a profound effect on how your company competes. In general, it causes great hardship for large, established firms because these firms tend to have major investments in older technology technology.. When a new technology substitutes for an old technology, these companies need to spend hundreds of millions and sometimes billions of dollars to replace their technology. technology. For example, the substitution of  VOIP technology for traditional circuit circuit switch telephone technologies technologies required the large, established telephone companies, like British Telecom and AT&T, to spend billions of dollars to replace their circuit switch technology with VOIP. VOIP. This substitution puts the established telecom companies at a disadvantage when they compete with small, new firms founded on the basis of the new technology technology.. Because technology substitution makes the investments of established firms obsolete, it is a very important part of the strategy that successful technology entrepreneurs use to compete with established firms, particularly in industries that face significant economies of scale. As new technology substitutes for old technology technology,, the scale economies at the businesses producing products based on the old technology erode, leading their costs to increase. When the established firms’ costs rise above those of the new firms, the established firms become uncompetitive. If the loss of  scale economies increases the cost structure of the established firms enough, new firms can drive the established firms out of business. For example, when the minimills entered the steel industry with a new technology that substituted for traditional steel casting, the integrated steel mills initially suffered only modest declines in the volume of steel that they produced. However, these modest declines in production volume set off a downward spiral that ultimately resulted in the bankruptcy of many integrated steel companies. Because economies of scale are very large in the steel industry, industry, the initially small substitution s ubstitution of new steel making technology that resulted in small declines in the volume of production at the integrated steel mills led to large increases in those mills’ cost of production. Those cost increases encouraged further substitution to the new

64

 

Technology Adoption and Diffusion

minimill technology, technology, which further reduced integrated steel mill production volumes, and further increased their production costs, creating a downward cycle for the integrated grat ed steel mill mills. s. While the steel mill example illustrates the value of substitution to the formulation effective technology at start-up companies, the formulation of  suchofa an strategy is difficult for strategy several reasons. First, substitution is very much affected by political actions to deter change taken by the providers of products based on the old technology. For example, the producers of glass windshields have taken action to keep auto insurers from authorizing repairs to cars made with plastic windshields because plastic could substitute for glass in windshield repair. Because auto insurers pay for most of the repairs to automobiles damaged in collisions, this political action has kept the substitution from occurring. Established companies that exploit the old technology usually have an advantage in the realm of political action because they employ large numbers of people whose jobs would be lost if the technology substitution occurs, creating a strong incentive for policy makers to support their efforts to deter substitution. Moreover, the mechanisms to discourage policy makers from taking actions that slow technology substitution involve lobbying and political influence, activities with which many entrepreneurs are unfamiliar. A second reason why a substitution-based technology strategy is difficult difficult to implement is that substitution is typically partial. As Figure 8. shows somewhat humorously, when one technology substitutes for another, the new technology usually does not completely replace the old technology. For example, the use of trucks did not completely substitute for the use of railroads to transport goods. Because substitution is usually partial, you need to figure out how much substitution will occur to determine if it will be enough for you to compete successfully against established firms who are exploiting the old technology. A final reason why a substitution-based substitution-based technology strategy is difficult difficult to pull off is that substitution takes very different amounts of time in different industries. For example, as Table Table 2 shows, it took synthetic rubber rubber 58 years to drive natural rubber down to only 10 percent of the market, but detergent soap took only 9 years to push natural soap down to the same percent market share. 83 As a result, with FIGURE 8

Complete Substitution Is Not Always Possible

Few technology products are ever complete substitutes for the products used before them. Source: New Yorker Magazine. 1975. The New York Album of Drawings, 1925–1975, 1925–1975 , New York: Penguin Books.

65

    

Technology Adoption and Diffusion

TABLE 2

Time to 90 Percent Time Substitution The time it takes for a new technology product to substitute for 90 percent of the uses of an old technology product varies widely,, making it difwidely ficult for technology strategists to plan for substitution.

SUBSTITUTES

Synthetic/natural rubber Synthetic/natural fibers Plastic/natural leather Margarine/butter Water/oil house paint Open hearth/Bessemer steel Sulfate/tree sapped turpentine Plastic/hardwood home floors Organic/inorganic pesticides Synthetic/natural tire fibers Plastics/metals in cars Animal/detergent soap

YEAR EARS S TO 90 PERCENT SUBSTITUTION

58 58 57 56 43 42 42 25 19 17 16 9

Source: Adapted from Girifalco, L. 1991. Dynamics of Technical Change. New York: Van Nostrand Reinhold.

soap, the companies introducing products based on the new technology were able to exploit the rapid deterioration of the cost structure of established firms in ways that companies introducing products based on synthetic rubber could not. Without accurate estimates of the time horizon of substitution, you cannot balance supply and demand dem and for products based bas ed on the new technology technology.. Take, for example, an automobile company that wants to develop electric vehicles as a substitute for gasoline-powered vehicles. If the company bets that the substitution will take 20 years and it occurs more quickly, the company will not have enough cars to satisfy demand and will allow competitors to enter. On the other hand, if the company bets that the substitution will take five years and it occurs more slowly, it will run out of  cash before the new technology begins to replace the old one. Established companies using the old technology also have to estimate the time horizon of substitution accurately because that horizon affects how quickly they need to transition to the new technology. Unfortunately, they often believe that this time horizon will be longer than it actually is, leaving them flat-footed in the face of  changing demand and competition. For instance, Kodak’s seniorfilm management ized that digital technology ultimately would replace traditional technologyrealtechnology, , but they assumed that this substitution would occur much more slowly than it did, leaving Kodak without adequate digital camera alternatives to its traditional products, and forcing it to undertake a painful restructuring. Sometimes established companies don’t wait for competitors to develop substitutes for their products and deliberately develop new products based on new technologies to substitute for their own products. Examples of these efforts include the development of 3G telephones in place of GSM phones, highdefinition DVDs in place of standard DVD technology, and digital televisions in place of analog ones. The deliberate development of new products based on new technologies requiress you to consider another important aspect of substitution, product cannibalrequire ization. This is a difficult and risky dimension of technology strategy. If you mismanage the transition to the new technology, your company will lose the sales of its old product without having a replacement for it, after having incurred the costs of  unnecessary product development. So, if you are going to substitute a new technology for an old one, you need to balance the advantages that you will derive from

66

 

Technology Adoption and Diffusion

deterring competitors who would otherwise introduce the new technology, technology, against the costs of product cannibalization.  Key Points

• Forecasting demand helps you to determine how much of your product to produce, to calculate the payback on your investment in product development, to project your costs, to set a pricing and advertising plan, and to formulate the right competitive strategy. • Forecasting demand is difficult because markets are dynamic, demand growth is nonlinear, nonlinear, and the accuracy of forecasts depends on the accuracy of estimates about the timing of customer adoption and the factors that affect the diffusion of technology. • Information diffusion models measure the functional form of diffusion, using data on the proportion of innovators and imitators in the customer population. • The Bass model is a tool to estimate the diffusion of technology products products that is  based on the rate of of adoption by innovators and imitators, imitators, and the size of the potential market. • Versions of the Bass model can incorporate the effects of the marketing mix,

• •

•

• •

changes in the size of the potential market, the effect of diffusion of complementary products, substitution by future generations of products, and competition  between products products with different different features. features. The Delphi Technique Technique is a methodology for gathering data from experts about technology trends. Product diffusion models measure the rate of diffusion, which depends on the costs and benefits of the new product, the amount of learning necessary to understand its benefits, the perceived risks of the product, the nature of the innovation, the characteristics of the target market, and the nature of the environment in which the adoption decision is being made. Substitution is the replacement of one technology for another that can achieve the same objective; it is an important part of technology strategy because it influences the competition between incumbent firms and new entrants. The implementation of a substitution strategy is difficult because substitution can  be multile multilevel, vel, face poli political tical oppo opposition sition,, be partia partial, l, and take a long or or a short short time. time. Companies sometimes deliberately develop new products to substitute for old products; this process is risky and can severely damage a firm’s competitive position if done incorrectly.

DISCUSSION QUESTIONS 1. What shapes can the adoption curve curve for new technology products take? What is the relationship  between the distribution distribution of of adopters and and the shape of the adoption curve? 2. Why do innovators, early adopters, early majority, majority, late majority, and laggards adopt products? What effects do differences in their motivations have on technology strategy? 3. What’s the difference difference between the adoption S-curve and the technology S-curve?

4. What is the typical pattern of diffusion diffusion of new technology products? What factors affect this pattern? How do these factors affect this pattern? 5. What are the similarities and differences differences between product and information diffusion models? 6. Is the Delphi Technique Technique better than than the Bass model at forecasting demand or vice versa? Why? 7. How should you manage technology substitution? substitution? Why do you make these recommendations?

67

 

Technology Adoption and Diffusion

KEY TERMS Adoption: The decision by customers to purchase a new product or service.

Innovators: (1) The group of customers that adopt a new technology product or service immediately

Bass Diffusion Model: A mathematica mathematicall tool for estimating the rate at which a new product or service diffuses to a potential market. Consumer Durables: Consumer products whose value is not exhausted in a year. Crossing the Chasm: The process of transitioning from the early adopters of a product to the early majority. Diffusion: The rate at which a new technology product or service is adopted by potential users. Early Adopters: The segment of the market that follows the innovators in the adoption of a new technology product or service. Early Majority: The segment of the market that adopts a product slightly before the market average. Horizontal Marketing Strategy: A stra strategy tegy of

after it is introduced; (2) customers who learn about new products from sources other than previous adopters. Laggards: The segment of the market that adopts new technology products or services last. Late Majority: The segment of the market that adopts new technology products or services after the early majority. Normally Distributed: A bell-shaped distributi distribution on in which a small portion of the population falls at each end, and the majority falls in the middle. Substitution: The replacement of one technology for another that can achieve the same objective. Take Off: The period in time when the S-curve of  adoption accelerates because of the transition

serving customers in multiple industries at the same time. Imitators: Customers who learn about new products from previous adopters.

from the early adopters adopters to the majority majority of the market. Vertical Marketing Strategy: A strategy of focusing on on serving customers in a single industry at a time.

PUTTING IDEAS

INTO PRACTICE

1. Technology Adoption Select a new technology product or service that you know well. Describe the distribution of adopters. (e.g., Is it normal or does it take another shape? If it takes another shape, what is that shape?) What proportion of  adopters fall into each group? Describe the characteristics of the different adopter groups? (e.g.,

3. Forecasting Technology Diffusion The purpose of  this exercise is for you to use the Bass model to predict the adoption of a new technology product, in this case the e-book. Your assignment is to develop a forecast for the diffusion of that product or service over the next 20 years. Assume that the maximum number of e-books that could ever be

innovators, early adopters, early majority, late majority, and laggards). Identify the factors that you think will influence adoption of this product or service by the different groups of adopters. Explain how to “cross the chasm” to gain adoption of the product or service by the majority of the market. 2. Information Diffusion Models Go to http://andorraweb.com/bass/index.=?show[exa mples]=1. Click on the diffusion models for the different products listed. What does the diffusion pattern look like for each of the products? (For example, are they S-shaped? Are they linear? Are they upward sloping curves?) Why are the diffusion patterns different? What conditions must be present to have each of the diffusion

sold in a year in the United States is 500 million. In addition, assume the rate of innovator adoption to  be 0.082 and and the rate of imitator adoption to to be 0.416. In an Excel spreadsheet, calculate the number of customers adopting annually, as well as the cumulative number of adopters in each year. Then plot the expected pattern of diffusion over the next 20 years. Now change the rate of innovator adoption to 0.20 and the rate of imitator adoption to 0.05. How does the adoption pattern change? What if you change the rate of innovator adoption to 0.05 and the rate of imitator adoption to 0.20? What happens if you leave the rate of imitator and innovator adoption the same but change the maximum market size never to be exceeded to 100 million? Why is the adoption pattern different under these different

patterns?

scenarios?

68

 

Technology Adoption and Diffusion

NOTES 1. http://www http://www.taipeitimes.c .taipeitimes.com/News/wo om/News/worldbiz/ rldbiz/ archives/2004/01/15/2003087725

24. Brown, Managing Managing the “S” curves of innovation. 25. Moo Moore, re, Crossing the Chasm.

2. http://www http://www.taipeitimes.c .taipeitimes.com/News/wo om/News/worldbiz/ rldbiz/ archives/2004/01/15/2003087725 3. http://www.usatoday.com/money/industries http://www.usatoday.com/money/industries/ / manufacturing/2004–01–13-kodak-cameras_x.htm 4. http://www http://www.boston.com/ .boston.com/business/arti business/articles/ cles/ 2005/07/21 5. Mahajan, V., V., and E. Muller. 1998. When is it worthwhile targeting the majority instead of the innovators in a new product launch? Journa  Journall of Marketin Marketingg Research Research, 35(4): 488–295. 6. Rogers, E. 1983. Diffusion of Innovations. New York: Free Press. 7. Calcula Calculated ted from information downloaded downloaded from http:// andorraweb.com/bass/i andorraw eb.com/bass/index.=?show[ex ndex.=?show[examples]=1. amples]=1. 8. Gatignon, H., J. Eliashberg, and T. T. Robertson. 1989. Modeling multinational diffusion patterns: An effi-

26. Afuah, A. 2003. Innovation Management. New York: Oxford University Press. 27. Moo Moore, re, Crossing the Chasm. 28. Mossberg, W. W. 2007. Nokia’s marriage to small computers still has its problems. Wall Street Journal, February 22: B1. 29. Lal, R. 2002. Documentum Documentum Inc., Harvard Business School Case, Number 9–502–026. 30. Moo Moore, re, Crossing the Chasm. 31. Ibid Ibid.. 32. Ibid Ibid.. 33. Rosch, W. W. 2004. New roofing shingles mimic nature. Cleveland Plain Dealer, June 17: D3. 34. Christiansen Christiansen,, C. 1997. The Innovator’s Dilemma. Cambridge: Harvard Business School Press. 35. Grant, P., P., and A. Latour. Latour. 2003. Battered telecoms face fac e

cient methodology. Marketing Science, 8(3): 231–247. 9. Moore, G. 1991. Crossing the Chasm. New York: Harper Collins. 10. Mohr Mohr,, J., S. Sengupta, and S. Slater. Slater. 2005. Marketing of High-Technology Products and Innovations (2nd edition). Upper Saddle River, NJ: Prentice Hall. 11. Vanac, M. 2007. Vision for the future. The Plain Dealer, February 14: C1, C3. 12. Brown, S., and V. V. Venkatesh Venkatesh.. 2003. Bringing nonadopters along: The challenge facing the PC industry. Communications of the ACM, 46(4): 76–80. 13. Burt, R. 1973. The differential differential impact of social integration on the participation in the diffusion of innovations. Social Science Research, 2: 125–144. 14. Schilling, M. M. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill.

new challenge: Internet calling. Wall Street Journal, October 9: A1, A9. 36. Christianse Christiansen, n, The Innovator’s Dilemma. 37. Ofek, E. 2005. Forecasting Forecasting the adoption of a new product. Harvard Business School Note, Number 9–505–062. 38. Van den Bulte, C., and S. Stremersch. 2004. Social contagion and income heterogeneity in new product diffusion: A meta-analyti meta-analyticc test. Marketing Science, 23(4): 530–544. 39. Mahajan, V., V., E. Muller, and F. Bass. 1990. New product diffusion models in marketing: A review Marke ting , and directions for research. Journal of Marketing 54(1): 1–26. 40. Cooper Cooper,, R., and E. Kleinschmid Kleinschmidt. t. 1987. New products: What separates winners from losers? Journal of 

15. Burgelma Burgelman, n, R., C. Christiansen, and S. Wheelwright. 2004. Strategic Management of Technology and Innovation . New York: McGraw-Hill Irwin. 16. Moo Moore, re, Crossing the Chasm. 17. Brown and Venkatesh, Venkatesh, Bringing non-adopters along. 18. Day Day,, G. 2000. Assessing future markets for new technologies. In G. Day and P. Schoemaker (eds.), Wharton on Managing Emerging Technologies Technologies . New York: John Wiley. 19. Brown and Venkatesh, Venkatesh, Bringing non-adopters along. 20. Roge Rogers, rs, Diffusion of Innovations. 21. Brown, R. 1992. Managing Managing the “S” curves of innovation. Journal of Consumer Marketing, 9(1): 61–72. 22. Ibid Ibid.. 23. Vara, V., V., and M. Mangalindan. 2006. 20 06. Web Web pioneers eBay and Amazon face a threat from older retailers.

Product Innovation Management, 4: 169–184. 41. Girfalco, L. 1991. Dynamics of Technological Change. New York: Van Nostrand Reinhold. 42. Dodson, J., and E. Muller. Muller. 1978. Models of new product diffusion through advertising and word of  mouth. Management Science, 24: 1568–1578. 43. Bass, F. F. 1969. A new product growth growth model for consumer durables. Management Science, 15: 215–227. 44. Ofek, Forecasting Forecasting the adoption of a new product. 45. Van den Bulte and Stremersch, Social contagion and income heterogeneity in new product diffusion. 46. Girf Girfalco alco,, Dynamics of Technological Change. 47. Roge Rogers, rs, Diffusion of Innovations. 48. Bass, A new product product growth model for consumer durables. 49. Ibid Ibid..

Wall Street Journal, November 16: A1, A12.

50. Ofek, Forecasting Forecasting the adoption of a new product.

69

 

Technology Adoption and Diffusion

51. Mahajan, Muller Muller,, and Bass, New product diffusion models in marketing. 52. Sultan F. F. 2001. Marketing Research for High Definition Television . Boston: Harvard Business School Publishing. 53. Kalish, S. 1985. 1985. A new product product adoption model model with pricing, advertising, and uncertainty. Management Science, 31(12): 1569–1585. 54. Mahajan, Muller Muller,, and Bass, New product diffusion models in marketing. 55. Ibid Ibid.. 56. Norton, J., and F. F. Bass. 1987. A diffusion theory theory model of adoption and substitution for successive generations of high technology products.  Management Science, 33 (9): 1069–1086. 57. Ofek, Forecasting Forecasting the adoption of a new product. 58. Ibid Ibid.. 59. Day Day,, Assessing future markets for new technologies. 60. Adapted from Ofek, Forecasting Forecasting the adoption of a new product. 61. Goldfisher Goldfisher,, K. 1992–1993. 1992–1993. Modified Delphi: A concept for new product forecasting.  Journal of Business Forecasting, 11(4): 10–11. 62. Day Day,, Assessing future markets for new technologies. 63. Wejnert, B. 2002. Integrating models of diffusion of  innovations: innovatio ns: A conceptual framework. framework. Annual Review of Sociology, 28: 297–326. 64. Tuzo, T. 1997. Asymmetric digital subscriber line: Prospects in 1997. Harvard Business School Case, Number SM-69. 65. Eisenmann, T., T., and F. Suarez. 2003. Symbian: Setting the mobility standard, Harvard Business School Case, Number 9–804–076. 66. Greenstein, S., and V. V. Stango. Forthcoming. The economics of standards and standardization. In

Technology and S. Shane (ed.), Blackwell Handbook on Technology Innovation Management. Oxford: Blackwell. 67. Roge Rogers, rs, Diffusion of Innovations. 68. Day Day,, Assessing future markets for new technologie technologies. s. 69. Ibid Ibid.. 70. Van den Bulte and Stremersch, Social contagion and income heterogeneity in new product diffusion. 71. Wejnert, Integrating models of diffusion of innovations. 72. Sandoval, G. 2005. Dueling Dueling DVD formats put buyers in the middle. The Plain Dealer, August 24: C3. 73. Roge Rogers, rs, Diffusion of Innovations. 74. Wejnert, Integrating models of diffusion of innovations. 75. Ibid Ibid.. 76. Rhoads, C. 2007. Web Web site to holy site: Israeli firm  broadcasts  broadca sts prayers for a fee. Wall Street Journal,  January 25: 25: B1–2. 77. Dhebar Dhebar,, A. 1995. The introduction introduction of FM radio (A), (B), and (C), Harvard Business School Teaching Teaching Note, Number 5–594–072. 78. Girf Girfalco alco,, Dynamics of Technological Change. 79. Three Mile Island Accident, http://en.wikipedia.org/ http://en.wikipedia.org/ wiki/Three_Mile_Island_accident. 80. Boudette, N. 2006. 2006. GM hopes engine of the future sells cars now. Wall Street Journal, November 29: B1, B2. 81. Roge Rogers, rs, Diffusion of Innovations. 82. Dranove, D., and N. Gandal. Gandal. 2003. The DVD-vs.DIVX standard war: Empirical evidence of network effects and preannouncement effects.  Journal of Economics and Management Strategy, 12(3): 363–386. 83. Girfalco, Dynamics of Technological Technological Change.

70

 

Sources of Innovation Learning Objectives Technological Opportunities: A Vignett Vignettee Introduction Sources of Opportunities Technological Change Political and Regulatory Change Social and Demographic Change Combination Locus of Innovation Research and Development Types of R&D Why Firms Invest in R&D

Costs of R&D Linking Research and Development Getting Down to Business: Investing to Create Opportunities Forms of Innovation Forms of Innovation and Technology Strategy Industry and the Nature of Innovation Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Define

a technological opportunity, opportunity, and explain why such su ch opportunities exist.

2. Spell

out how technological, political, regulatory, regulatory, social, and demographic changes generate opportunities for technological innovation.

3. Identify

the different loci of innovation, and explain how different institutions contribute to technological innovation.

4. Describe

the roles played by the public sector in the national innovation system, and explain how universities help private firms to innovate.

5. Identify

the different components of R&D, and figure out the pros and cons of investing in R&D.

6. Describe

the different ways that organizations link R&D activities, and account for the trade-offs between the different approaches.

From Chapter 4 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 71

 

Sources of Innovation

7. List

the different forms of innovation, and explain why technological innovation takes these different forms.

8. Interpret

the effects of industry on the forms of innovation, and figure out the right strategy for exploiting the different forms of innovation.

Technological Opportunities: A Vignette1  Howard Becker, Becker, the CEO of a Mayfield Heights, Ohio, company called Comet Video Technologies Technologies LLC, took advantage of two sources of innovation in developing his company’s first product, Comet Vision. This device can send streaming video to other computers, cell phones, or personal digital assistants when it is connected to a video camera, a computer, and the Internet. Becker is targeting this product at parents of older children who want to know what the children are doing when they are home alone. By attaching up to six video cameras to a computer, parents can see what is happening inside their homes by accessing their office computers, cell phones, or personal digital assistants. Becker’s company took advantage of an important technological change, the development of a wireless compression technology that allows video to be streamed to telephones and personal digital assistants at a low bandwidth. Most video is transmitted electronically through mathematical algorithms, such as JPEG and MPEG. However, these algorithms require a high bandwidth. Comet Video Technologies LLC developed a new algorithm that transmits video at a low bandwidth, making possible video transmission to a wider variety of electronic devices. The company also took advantage of an important social change, the concern of working parents about what their children are doing when the children are not with them. With a greater number of children who are home alone after school, getting into trouble, many parents want a way to keep an eye on what their children are doing when they cannot be at home. Comet Vision shows how entrepreneurs—and managers of established companies—often develop technological innovations in response to social and technological changes that create opportunities for new products and services. It also demonstrates how many opportunities result from a combination of different changes, in this case social and technological.

I NTRODUCTION Opportunities to create new products and processes emerge from change—in technology, in politics and regulation, and in social and demographic factors. Effective technology strategists either create or identify those opportunities and figure out the  best way to take advantage of them—by creating new products, establis establishing hing new production processes, organizing businesses in new ways, introducing new inputs, or tapping new markets.

72

 

Sources of Innovation

The first section of the chapter examines the sources of opportunities, focusing on technological, political-regulatory, and social-demographic changes. The second section focuses on the entities that engage in technology development and explains how each of them contributes to the innovation process. The third section explains why firms invest in research and development, despite an inability to capture all of  the returns to investment in it. The final section examines the different forms that innovation can take, and the key relationships between the types of changes that occur and the form of innovation.

S O U R C ES ES

O F   O PPORTUNITIES

The technological innovation that firms undertake is often triggered by some kind of  change, whether that change is the result of organized human action or is accidental. For example, the invention of the laser was a source of opportunity because that invention made it possible to develop a new product for storing music (the compact disk), a new product for ringing up groceries (the checkout scanner), and a new service for correcting vision problems (laser eye surgery) among other things. In the absence of the invention of the laser, the creation of these products and services would not have been possible. Researchers have identified three major sources of opportunity for innovation: technological change, political and regulatory change, and social and demographic change, which are discussed next.

Technological Change As you might expect, technological change is one of the most important triggers of technological innovation, largely because technological change allows people to do things that could not be done before or only could be done in a less efficient manner. 2 Take, for example, the invention of the computer software behind e-mail. This software made it possible to communicate in ways that people find “better” for many purposes than advance communicating by telephone, fax, or letter. technological change here—the in computer software—opened upThe an opportunity for innovation: the creation of a new communication product called e-mail. It’s important for you to understand the connection between technological change and opportunities for innovation. Unfortunately, it isn’t a straightforward, one-to-one relationship. Many technological changes do not make any innovations possible, while others generate a multitude of opportunities. For example, less than 5 percent of all newly patented inventions ever lead to a commercial product product or service.3 On the other hand, some technological changes, like the microchip, have led to the creation of a very large number of new products.4 Sometimes, the introduction of a new technology creates demand for new products to counteract its negative side effects, as well as to exploit its benefits. For instance, the development of the Internet has helped companies by allowing their employees to get easier access to information that facilitates purchasing, evaluation of potential products and employees, information search, and a variety of other uses. However,, Internet connections at the office also allow employees access to pornographic, However

73

 

Sources of Innovation

gambling, hateful, and other types of Internet sites that could potentially open up companies to liability liability..5 Thus, the development of the Internet has created the opportunity for companies to produce Web-filtering software that allows Internet search while blocking objectionable sites. Technological change also does not immediately lead to opportunities for innovation, making the time from invention to innovation uncertain and potentially very long. Take, for example, the case of the photocopy machine. Chester Carlson invented xerography in 1937, but it was not until 1949 that the first Xerox photocopying machines were introduced to the market. Many times this delay occurs because complementary technologies, or technologies that are used along with the focal technology, need to be invented before an innovation can be developed. For example, the development of high-speed radio devices, like cellular telephones, first required advances in complementary metal oxide semiconductor (CMOS) technology, which made it possible to use CMOS chips rather than just using ones made from gallium arsenide or silicon germanium. As a result, cellular telephones were not immediately developed once they were technically possible.6 What makes predicting the timing of the relationship between technological change and opportunities for innovation so complicated is that technological change is rarely linear. Take Take fuel cells as an example. The invention of fuel cells did not immediately open up the opportunity to change automobile engines from internal combustion to fuel cells. Why? Because the cost effectiveness of producing fuel cells has grown very slowly. An automobile needs an engine that generates 100 kilowatts of power. Because it costs about $5,000 $5,000 to produce the power power plant of a car, car, that’s a price of about $50 per kilowatt. However, However, fuel cells currently cost about $3,000 $3,000 per kilowatt. kilowatt. Consequently Consequently,, right now, now, there is no commercial

FIGURE 1

Technological Change Is an Important Source of Innovation

Even the problems created by the introduction of new technologies create opportunities for other innovations. http://www.speedbump.com, .speedbump.com, January 23, 2007. Source: http://www

74

 

Sources of Innovation

opportunity to build cars powered by fuel cells. As improvements in the technology lower the per kilowatt cost of fuel cells, the opportunity for people to make fuel cell cell cars should should open up.

Important Attributes of Technological Change  Because there is not a simple, one-to-one relationship between technological change and innovation, technology strategists need to evaluate technological changes to determine if they are likely to make valuable innovations emerge in the future. So what attributes of technological change are associated with subsequent innovation? The magnitude of the technological change is one important factor. Changes of  greater magnitude open up more opportunities for innovation because they allow new technology to be used in more ways than changes of small magnitude permit. Take, for instance, the creation of a new type of electrical circuit. If that new circuit is only 10 percent faster than an older one, it will replace the older circuit in only a small number of products. Only those products in which a 10 percent improvement would exceed the cost of the change will use the new circuit. In contrast, if the new circuit is 500 percent faster than the older one, its benefits will exceed the cost of the change in a much larger range of products. Then there is the generality of the change. Some new technologies, like genetic engineering or nanotechnology nanotechnology,, are general purpose technologies. They make possi ble the creat creation ion of a wide range of new prod products. ucts. For inst instance ance,, the inve inventio ntion n of  genetic engineering has led to the creation of new products in human and animal health-care, agriculture, and industrial chemicals, while the invention of nanotechnology has made possible new products in televisions, sensors, transistors, batteries, resins, and clothing.7 General purpose technologies are particularly important to innovation by new firms because these technologies permit strategic flexibility. If a new company is exploiting a general purpose technology and one market application proves to be inappropriate, the company can shift to other applications, minimizing the likelihood that the company will reach a dead end and will have to shutt dow shu down. n. General purpose technologies do not provide the same benefit to established firms because these firms see general purpose technologies through the lens of their existing businesses. Consequently, they their can rarely take advantage of the flexibility that these technologies provide, given unwillingness to pursue technology development when the value of a technology lies lie s in another industry. The commercial viability of the change also matters. Some new technologies do not offer much commercial potential—at least not for a long time. For instance, the space shuttle is a very large change over alternatives for getting into space because it saves a huge amount of money over rockets that cannot be reused. However, the commercial benefits of the space shuttle are rather limited because there are only a few commercial applications to which it can be put. Of course, you need to be careful in thinking about the commercial viability of the change. Sometimes, a change will lead to a large number of commercial outcomes, but people don’t recognize that immediately. This is what happened with the laser. At the time it was invented, no one thought that it would have any commercial use; but just the opposite happened. The laser turned out to have commercial viability in a wide range of areas. The lesson here is that the commercial viability of a technological change affects subsequent innovation,  but it is ofte n very v ery diff difficul icultt for ent entrepre repreneu neurs rs and mana manager gerss to see tho those se opp opporortunities at the time that the change occurs.

75

 

Sources of Innovation

Political and Regulatory Change Political and regulatory changes also make innovation possible in a variety of ways. Regulation sometimes makes innovation possible by providing subsidies that pass off the cost of innovation to the government, thus reducing the cost of innovation to the companies undertaking it.8 (A subsidy is a payment payment by the government government that makes up the difference between what customers will pay for a product or service and the cost of producing it.) For example, the governments of several European countries created Airbus, an aerospace rival to Boeing, by subsidizing the cost of  R&D investments by that company. Similarly, changes in federal tax credits for hybrid vehicles in the United States have increased the incentives for auto companies to produce those cars. Another way that political and regulatory change can make innovation possible is— perhaps paradoxically—by paradoxically—by spurring firms to come up with new products and processes to solve problems created by regulation.9 For example, European Union restrictions on the importation of hazardous materials led Corning to develop a new type of glass for liquid crystal displays that is free of heavy metals, and can be used in televisions and laptop screens sold in Europe.10 Similarly, the Sarbanes-Oxley Act, which requires publicly traded companies in the United States to ensure that their accounting and finance systems can detect and deter fraud, has created an opportunity for software companies to create new products to store audit-related materials electronically. 11 Political and regulatory change can also make innovation possible by spurring competition between firms. As firms seek to generate competitive advantages, they often innovate. For instance, the deregulation of telecommunications led firms in that industry to compete to introduce less expensive ways to transmit voice and data that gave them a price advantage over their competitors. 12 Similarly, regulations in France that required the national telephone company to make its phone lines available to other Internet service providers has spurred competition that has led those providers to offer download speeds over 20 times faster than in the United States, where federal courts have blocked regulators from implementing similar policies. 13 Political and regulatory change can make innovation possible by providing access to resources that permit the development of new products and services. This

FIGURE 2

Political and Regulatory Change

Located in Rear Filler Panel of Passenger Cars

Located in Cargo Floor of Station Wagons/Vans

The child’s car seat shown here is i s a product that resulted from changes in government regulations about such things as the mechanism to latch the seat to the car and the stability of the base. http://www.nhtsa.dot.gov/CPS/safetycheck/ .nhtsa.dot.gov/CPS/safetycheck/ Source: National Highway Traffic Safety Administration, http://www TypeSeats/index.htm.

76

 

Sources of Innovation

often occurs when the government controls the access to key resources and changes the regulations about how companies can access them. For instance, in the United States, the Federal Communications Commission’s decision in 1997 to open up 300 MHz of the radio spectrum in the 5 GHz frequency for short-range highspeed digital communications created opportunities for companies to develop an innovative product called wireless local area network (LAN).14

Social and Demographic Change Another important category of change that generates opportunities for innovation is social and demographic change, which opens up opportunities by altering people’s preferences and by creating demand for products that had not existed  before. For Fo r instance, ins tance, the shift shi ft of women into the t he workforce, workfo rce, and the corresponding cor responding increase in demand for speed in food preparation, created the opportunity to introduce many types of frozen food into the marketplace. Similarly, as Figure 3 shows, a FIGURE 3 Market Share of US Internet Visits

Social and Demographic

0.55%

Change

0.50%

0.52% 0.45% 0.40% 0.35% 0.32% 0.29%

0.30% 0.25%

0.20%

0.20% 0.15% 0.10% 0.05% 0%

   l   u    l   g    g    p   p   p   c    t   c    t   a  r    M  a  r   A  p  r   A  p  r   A  p  r   a  y   a  y   J  u  n   J  u  n   J   u   u   A  u   S  e   S  e   S  e   O   J    A   O    M              M    M                     8   2       1    5   9   4   8   0   4    5   1  9   0  2   1  6   3  0   1   2    7   1   2   0   2   0   0  4   1  8   0   1   2   1  3   2

www.facebook.com

www.youtube.com

www.photobucket.com

www.wikipedia.com

Chart of the weekly all sites market share in “All Categories”, based on visits. Time periods represented with broken lines indicate insufficient data. Generated on: 11/08/06. Copyright Copyright 2006 (C) “Hitwise Pty Ltd”.

Social trends toward the use of technology to keep in touch friends created the opportunity for social networking Web sites like Facebook, which, as thewith figure shows, has become one of the more heavily viewed Internet sites. /weblogs.hitwise.com/leeann-prescott/2006/11/social_networking_sites_recove.html. 1/social_networking_sites_recove.html. Source: http://weblogs.hitwise.com/leeann-prescott/2006/1

77

 

Sources of Innovation

desire of people to be more closely connected and in greater touch with their friends led to the development of social networking Web sites, like Facebook and MySpace, which have quickly become some of the most used Internet sites. One important type of opportunity-generating social and demographic change is a social trend, which leads potential customers to change their needs and preferences. For example, the opportunity to produce deodorant was the result of a social trend that led people to believe that body odor was offensive. While there is actually no medical or health need for people to mask their body odor, the trend in believing that body odor is offensive opened up the opportunity for companies to create products that mask that odor. Demographic trends are another important source of opportunity to introduce innovative new products and services. For example, as birth rates decline and people live longer, the population of many developed countries is aging. This demographic trend makes it possible to introduce products and services targeted at the elderly, for which sufficient demand did not exist 25 years ago. In Japan, Toyota is responding to this demographic trend by producing Welcabs—specially Welcabs—specially designed cars with wheelchair ramps and special seats that make it easier for the elderly to get in and out. 15 A third type of opportunity-creating opportunity-creating social or demographic demographic change is a shift in perception. Sometimes, an opportunity to innovate occurs because people perceive things in a new way way.. Take, for example, the use of vaporized hydrogen peroxide to clean facilities contaminated with anthrax and other biological agents. Steris Corporation, a Cleveland, Ohio, medical device firm, adapted this technology,, which it had originally developed for decontaminating pharmaceutechnology tical clean rooms, to work against biological agents. Why? The anthrax terrorism scare in the Hart Senate Office Building and U.S. post offices changed people’s perception about the need for government offices to be clean of biological contaminants.16

Combination The previous discussion of the effects of technological, political, p olitical, regulatory, social, and demographic onHowever, innovation described sources of opportunity separately from onechange another. this approachthese is purely an artifact of exposition. (The relationship between the change and the creation of opportunities to innovate would not be as clear if the changes were not discussed separately.) In reality,, opportunities for innovation are often the result of many different types of  reality changes. Take, for example, the efforts of companies like Net Nanny and Cyber Patrol to introduce new products to protect children against harmful Internet content. The need to protect children against certain Internet content came about both because of  a technological change (the invention of the Internet) and because of a social trend (the tendency for children to be home alone after school). Similarly,, in the United States, a start-up company called iFly has developed an Similarly air taxi service that will let corporate travelers fly in small planes between local airports, much like a limousine service takes passengers between locations within a city. The opportunity for iFly’s innovation came about as a result of both technological, and political and regulatory changes. On the technological side, the development of four-person micro jets with fuel efficient engines, something that previously had not been technically feasible, was an important change that made

78

 

Sources of Innovation

this innovation possible. On the political and regulatory side, the time-consuming post-September 11 security procedures procedures at U.S. airports was an important source of  17 the opportunity.  Key Points • Technological, regulatory, regulatory, political, social, and demographic changes are all sources of innovation and can operate separately or in combination. • Technological changes vary in their magnitude, generality, and commercial viability,, all of which influence opportunities for innovation. viability • Predicting the relationship between technological change and the opportunity for innovation is difficult because the relationship is not always one-to-one, leads to the creation of additional opportunities, and because the effect of technological change on the creation of opportunity is rarely immediate. • Political and regulatory change affects opportunities for innovation by changing the costs of, or rewards to, innovation; by spurring firms to respond to problems created by regulation; by providing potential innovators with access to resources and subsidies; and by spurring competition

 between  betwe en firms. fi rms. • Social and demographic changes affect opportunities for innovation by altering people’s preferences and by creating demand for things that had not existed before.

LO CUS

O F  I NNOVATION

A variety of different entities undertake technological technological innovation, including companies, individuals, universities, and government agencies. Of these four sources, companies account for the most technological innovations. Sometimes company innovation occurs because businesses are trying to come up with new productss or services to sell to customers or to improve their processes for making product and distributing thoseinnovations products or services. Other times, develop technological because no supplier offershowever, a productcompanies or service 18 that they need. For example, Texas Instruments, a customer of semiconductor manufacturing equipment, developed the planar process for making semiconductors on a silicon substrate because no other companies could provide it with that technology techno logy..19 Another important source of technological innovation is the government. Federal government agencies in many countries come up with their own technological innovations. For instance, researchers at Los Alamos National Laboratory developed a radio frequency identification system for locating nuclear materials, which led to the development of automated toll payment systems, like the E-ZPass.20 The government also helps develop technological innovation by sponsoring  basic research undertaken by universities and industry,21 particularly in the life sciences and engineering (see Figure 4). For instance, the Defense Advanced Research Projects Agency (DARPA) of the Defense Department has funded much of the basic research in computer science in the United States, including basic research on the Internet that began in the 1960s.22

79

 

Sources of Innovation

FIGURE 4

Share of Federal Funding of  Different Technical

Life Sciences 2%

Fields

2% Psychology

17%

Physical Sciences Environmental Sciences 5%

55%

Mathematics and Computer Sciences

7% Engineering Social Sciences

10% 2%

Other Sciences, nec

Federal funding of research is not equal across fields; life sciences account for over half of all federal funding of R&D in the United States. Source: National Science Foundation/Division of Science Resources Statistics, Survey of Federal Funds for Research and Development, Development, Fiscal Years 2002, 2003, and 2004.

The government also sponsors corporate R&D. By paying for some of the R&D costs of private companies, governments increase the willingness of firms to develop technological innovations.23 For instance, General Motors Corp., Ford Motor Co., Hyundai Motor Co., and DaimlerChrysler are all recipients of U.S. Department of  Energy funding under a program to develop vehicles that run on fuel cells, rather than on gasoline.24 To help industry developto new technology technology, , in 1986, the U.S. federal government created a new arrangement allow joint business-government research efforts called Cooperative Research and Development Agreements (CRADA). Under these agreements, government laboratories can participate in cooperative research with industry with the government matching the financial contributions of industry with contributions of personnel and equipment. 25 As Figure Figure 5 shows shows,, while these agreements were once very popular, they are now being created at a much lower rate. In addition to its funding of research, the government also supports technological innovation by serving as a lead customer, and paying a high price, for the initial versions of new technology products.26 The classic example of a government agency that serves as a lead customer is NASA, the U.S. space agency, which pays a high price for the initial versions of many new technology products, which then are developed for other customers. Individuals are a third locus of innovation. While individuals once accounted for a much larger number of technological innovations than they do today, they are still an important locus, particularly for innovations that lead to the creation of new

80

 

Sources of Innovation

FIGURE 5 140

CRADA Alliances

120 100      s      e      c      n      a       i       l       l       A

80 60 40 20 0 1985

1987

1989

1991

1993

1995

1997

1999

2001

2003

This figure shows that the number of new CRADA alliances formed each year grew substantially from the mid 1980s to the mid 1990s, but then returned to the level of the late 1980s.

Source: Science and Engineering Indicators, 2006, http://www.nsf.gov/statistics/seind06/c4/fig04–18.htm.

companies. For example, Steve Jobs and Steve Wozniak created the Apple I personal computer in the garage of Steve Jobs’s parents before starting a company, making the original Apple computer an individual innovation. 27 Academic institutions are a fourth important locus of innovation. In fact, one study showed that 11 11 percent of all new products products and 9 percent percent of all new industrial processes would not have been developed in a timely manner if it were not for academic research. 28 Moreo Moreover ver,, Figure 6 shows that that the amount of  research conducted at universities is growing. Since the 1960s, two forces have led to the increasing amount of real university research funding: the financing of  defense-related R&D during the cold war and later the war on terrorism; and the financing of biomedical research and the rise of National ofbecause Health (NIH) funding. In 1998, academic research funding further Institutes accelerated economic growth and increases in tax revenues led to a federal budget surplus and a relaxation of restrictions on federal spending on university research. 29 R&D spending in industry has declined some in recent years (but then rebounded)  because  becau se of the move movement ment of much R&D activ activity ity out of the Unit United ed State Statess to t o other o ther countries, a shift of many large American companies out of basic research, the  bur st stin ing g of th thee in info form rmati ati on te techn chn ol olog ogy y bub bl blee in 200 0, an and d th thee de decre cre as asin ing g demand of the federal government for military-related innovations. Universities often conduct basic scientific research that leads to applied R&D, which results in technological innovation. For example, Ben & Jerry’s has sought to  build an acoustic freezer based on the results of U.S. Navy funded research at Penn State that uses sound waves instead of chlorofluorocarbons and hydrofluorocarbons as a cooling mechanism. In addition, many universities license the technological innovations developed by their faculty, staff, and students to the private sector and share the royalties from those licenses with the inventors. 30 While this activity has been

81

 

Sources of Innovation

FIGURE 6

The Rise of Academic Research

250,000

200,000

  s   r   a    l    l   o    D    0 150,000    0    0    2    f   o   s 100,000   n   o    i    l    i    M

50,000

0.00         3         4       5         6       7         8         9         0         1         2         3         4       5         6       7         8         9         0         1         2         3         4       5         6       7         8         9         0         1         2         3         4       5         6       7         8         9         0         1         2         3         4       5         6       7         8         9         0         1         2         3         4       5         6        5       5       5       5       5       5       5         6         6         6         6         6         6         6         6         6         6       7       7       7       7       7       7       7       7       7       7         8         8         8         8         8         8         8         8         8         8         9         9         9         9         9         9         9         9         9         9         0         0         0         0         0         0         0         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         9         0         0         0         0         0         0         0         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         1         2         2         2         2         2         2         2

Year

Federal

Industry

Colleges

Nonprofit

The amount of research undertaken at universities and colleges has grown dramatically over the past five decades, and it is now above the level undertaken by the federal government. Source: Created from data contained in National Science Foundation, Division of Science Resources Statistics, National Patterns of  R&D Resources, annual series. See appendix table B-1.

occurring since thethe 1920s, the number of passed universities engaged ingave it has grown dramatically since Bayh-Dole Act was in 1980. This law universities the rights to federally funded inventions, which account for about two-thirds of all university inventions, thus creating a strong incentive for universities to license their inventions to industry. Gatorade, the sports beverage, and Taxol, the cancer drug, are both examples of new products developed from licenses to university inventions. Universities also train students, who bring cutting-edge knowledge to the businesses that hire them. 31 This role of universities is so important that Intel funds semiconductor researchers at several major U.S. universities just so that they can hire the students of those researchers as development engineers. 32 Because universities are an important locus of technological innovation, many companies develop strategies to leverage the value of university research. For instance, Intel has established several small research labs just outside the campuses of  the University of California at Berkeley, Berkeley, Carnegie Mellon University University,, the University of  Washington, and Cambridge University. In these labs, 20 Intel and 20 university researchers work side-by-side to learn about university research and figure out how to transfer it to Intel.33

82

 

Sources of Innovation

All of the previous not withstanding, universities are a more important locus of  innovation for some technologies than others. For instance, university research is particularly important in semiconductors, measuring and controlling devices, pulp and paper, drugs, petroleum refining, aircraft, surgical and medical medi cal instruments, and computing.34 And the academic fields of chemistry, computer science, electrical engineering, materials science, and mechanical engineering are more important than other disciplines to technological innovation. 35 Therefore, the degree to which technology strategists focus on universities as a locus of innovation depends a great deal about what businesses their companies are in.  Key Points

• Business undertakes most of the R&D conducted in the United States. • Individuals conduct much less of the technological innovation than they once did. • The government plays an important role in the technology innovation process  by conducting research, paying for research done by others, and by serving as a lead customer. • Universities help firms innovate by training students, by conducting research, and by licensing technology technol ogy developed by their thei r faculty, faculty, staff, and students.

R ESEARCH

AND   D EVELOPMENT

As the previous section indicated, not all innovation is reactive. Companies can, and do, create innovations through deliberate investment in R&D efforts designed to create new products and services, and new processes for providing them. The implicit model behind corporate R&D is shown in Figure 7. Investments are made in R&D to create technological inventions. Those inventions make possible innovations, which, in turn, are embodied in new products and services that meet previously unmet needs or allow existing products and services to be produced more efficiently. All of this is done with the goal of capturing the commercial value generated from the effort.

Types of R&D R&D is composed of three different activities: Basic research, applied research, and development. Basic research is the effort to understand the technical or scientific

FIGURE 7

Model Underlying Investment in R&D

Investment in R&D

Technological Invention

Technological Innovation

New or Improved Products or Processes

Value Created and Captured

This figure shows the process model underlying investment in R&D; companies invest in R&D to create technological inventions that make possible innovations, which they can use to produce new products and services whose commercial value they try to capture.

83

 

Sources of Innovation

principles in a field. Usually, it is undertaken without commercial goals in mind. For example, scientists conducted basic research in physics that led to the discovery of the transistor.36 Applied research is the effort to understand technical or scientific principles with a specific commercial goal in mind. For example, scientists and engineers conducted applied research on the transistor to develop a new type of o f radio receiver that 37 could be used with it. Development is the effort to use technical knowledge to produce something of  commercial use. For example, engineers used knowledge of transistors and radio receivers to design a transistor radio that could be sold to customers.

Why Firms Invest in R&D Of the approximately $300 billion spent annually on R&D in the United States, industry accounts for approximately 71 percent, the federal government accounts for 8 percent, colleges and universities account for 16 percent, and nonprofit organizations account for 5 percent.38 However,, as Table However Table 1 shows, private sector R&D expenditure is not evenly distributed across firms. Large firms account for most of the R&D spending (though not as a percentage of sales); and, even among them, R&D expenditure is highly skewed,

TABLE 1

Top 20 R&DSpending Corporations in 2003 R&D spending by companies is highly concentrated in a small number of businesses;  just thr three ee compan companies— ies— Microsoft, Ford, and Pfizer—accounted for 11 percent of all U.S. R&D.

R&D (MILLIONS ) 2003

R&D AS PERCE ERCENT NT OF SALES 2003

Microsoft (United States)

7,779

21.1

Ford Motor (United States) Pfizer (United States) DaimlerChrysler (Germany) Toyota Motor (Japan) Siemens (Germany) General Motors (United States) Matsushita Electric Industrial (Japan) International Business Machines (United States) GlaxoSmithKline (United Kingdom)  Johnson & Johnson (United States) Sony (Japan) Nokia (Finland)

7,500 7,131 6,689 6,210 6,084 5,700 5,272

4.6 15.8 4.1 3.9 6.8 3.1 7.7

5,068

5.7

4,910

13.0

4,684

11.2

4,683 4,514

6.9 12.8

Intel (United States) Volkswagen (Germany) Honda Motor (Japan) Motorola (United States) Novartis (Switzerland) Roche Holding (Switzerland) Hewlett-Packard (United States)

4,360 4,233 4,086 3,771 3,756 3,694 3,652

14.5 4.0 5.5 13.9 15.1 15.3 5 .0

COMPANY (COUNTRY)

Source: Adapted from information in Science and Engineering Indicators 2006, http://www.nsf.gov/ statistics/seind06/c4/tt04–06.htm.

84

 

Sources of Innovation

TABLE 2

R&D Intensity of  Different Industries There is wide variation in the R&D intensity of the different industry sectors, with professional, scientific, and technical services investing over 10 percent of  their sales in R&D, and utilities investing only 0.10 percent.

INDUSTRY

R&D INTENSITY

Beverage and tobacco products Chemicals

0.5 5.6

Computer and electronic products Construction Electrical equipment, appliances, and components

9.3 1.2 2.2

Fabricated metal products

1.5

Finance, insurance, and real estate Food Furniture and related products Health-care services Information Machinery Management of companies and enterprises Mining, extraction, and support activities Miscellaneous manufacturing Nonmetallic mineral products

0.3 0.6 0.8 2.2 5.7 4.2 4.1 3.3 5.6 1.0

Other nonmanufacturing

4.9

Paper, printing, and support activities Petroleum and coal products Plastics and rubber products

1.1 0.3 2.1

Primary metals Professional, scientific, and technical services Retail trade

0.7 10.2 0.8

Textiles, apparel, and leather Transportation and warehousing Transportation equipment Utilities

1.0 0.4 2.7 0.1

Wholesale trade Wood products

3.6 0.7

All industries

3.2

http://www.nsf.gov/statistics/ ov/statistics/ Source: Adapted from Science and Engineering Indicators, 2006, http://www.nsf.g seind06/append/c4/at04–22.xls.

with only eight companies accounting for one-quarter of all R&D conducted by the private sector in the United States. 39 Moreover,, as Table Moreover Table 2 shows, private sector R&D expenditures vary greatly across industries, with some industries spending spe nding a much higher percentage of their sales on this activity than others. This variation across companies and industries in investment in R&D suggests that companies have both reasons to invest and reasons not to invest in R&D, and the relative balance between the two depends on the industry and the firm. So why do firms invest in R&D? There are basically five reasons: 1. To create new technologies that can serve as the basis for new products and serse rvices.40 For example, DuPont invested in research in polymers to create nylon, 41 and Monsanto invested in research in genetic engineering to create new agricultural pest control products.42

85

 

Sources of Innovation

2. To develop products to replace those threatened by substitutes. For example, AMP invested in research to replace its coaxial cable products products because the company’s managers believed that fiber optics would substitute for coaxial cable and make its products obsolete. 43 3. To differentiate products from those of competitors. competitor s. For example, many machine tool makers in the United States and Germany invest in R&D to develop versions of their products that have features not present in the products made by their lower-cost Chinese competitors. 4. To create strong intellectual property positions by making fundamental technological discoveries on which pioneering patents can be obtained. For example, investment in R&D allowed Texas Instruments to develop the pioneering patent for the integrated circuit, which positioned the company as a leader in semiconductor technology for many decades. 5. To create absorptive capacity, which is the capability to recognize and use knowledge from elsewhere. This capacity allows companies to use information from customers, suppliers, and even competitors to develop new products and services, or new ways of making them. 44 For example, IBM conducts R&D on X-ray lithography even though it doesn’t use that technology to make semiconductors so that it can recognize and take advantage of any developments made outside the company to use that technology for the purpose of making semiconductors.45

Costs of R&D While investing in R&D provides the benefits just described, they come at a cost. First, investments in basic research are uncertain, leading many investments to fail to generate a positive return. R&D projects often require years of investment before they reach the point of commercial viability, and only a small portion of the projects ever yield commercial outcomes. For instance, to develop a single cholesterol drug, Pfizer Inc. had to examine 400,000 chemicals, looking for possible drug leads. It then had to test those drug leads on animals to determine efficacy. efficacy. Following that, Pfizer had to go through several stages of FDA approval before before it had a cholesterol drug that it could sell on the market. Until the company hadthe anhundreds approved drug on the market, it could not capture any value from of cholesterol millions of dollars that it 46 had invested in the development of the drug. As a result, almost all of the financial returns from R&D projects are generated by a handful of projects, many years after the investments are first made, leaving average returns poor. For example, investors in public biotechnology companies have put more than $100 billion into those companies since they were first founded, but the cumulative returns for all public biotechnology companies since the formation of the industry has been negative $40 billion. 47 Second, investments in basic research often lock companies into strategies that are hard to change. Because the returns to investment in R&D only occur after years of investment, companies have to stick with particular development paths over long periods of time to benefit financially from investments in basic research. Unless a company can accurately predict the direction of technological change far into the future, it might become locked into the wrong strategic focus by selecting the wrong research projects and hiring the wrong researchers. For example, to conduct basic research on semiconductors, IBM must hire researchers with expertise on particular classes of molecules. These researchers

86

 

Sources of Innovation

cannot easily change their research focus to other classes of molecules, let alone to investigate Internet search engines. So if IBM is wrong about the direction of technological development, it will be stuck with semiconductor researchers whose work leads to dead ends, while lacking the expertise necessary to develop technology in a growth area like search engines. 48 Third, the returns from investments in R&D are difficult to appropriate. (Appropriability is the degree to which the value that results from an investment in the development of an innovation can be captured by the company that made the investment.) This lack of appropriability stems from two factors: • R&D has positive externalities, which make it possible for companies that did not pay for the R&D to benefit from the technological innovations of other firms.49 For example, Xerox’s discoveries of the Ethernet, graphical user interface, and computer mouse helped other firms figure out how to develop products from these technologies. • Knowledge leaks, or spills over, from one firm to another. For example, when one company hires employees empl oyees away from another company, those employees  bring with them knowledge of what their prior employer emp loyer was doing, creating 50 knowledge spillovers. (As a result, some companies, such as Samsung51 deliberately try to capture knowledge from other companies by systematically hiring their employees, particularly in technical areas that are new to the hiring company.52)

Linking Research and Development Part of technology strategy is concerned with linking research and development. The failure to link basic research with subsequent development will keep you from earning financial returns on your investments in basic research. For example, Xerox lost many billions of dollars of potential returns on its investment in basic research on the computer mouse, the graphical user interface, and the Ethernet because the company had no mechanism for linking these inventions to the activities of its operating units, which controlled development projects. However, creating effective links between research and development is not easy  because the former is undermined when it is closely clos ely tied to the latter. If researchers  become too heavily involved in short-term s hort-term problem solving on behalf b ehalf of o f operating ope rating units, then they do not have time for research projects. 53 Consequently, many

TABLE 3

The Benefits and Costs of R&D There are advantages and disadvantages to conducting basic research, which firms must balance when deciding whether or

not to undertake it.

BENEFITS

COSTS

Creates products and services that meet customer needs

Is difficult to appropriate the returns

Replaces products and services threatened by substitutes

Is highly uncertain and does not yield high average financial returns

Develops the absorptive capacity to recognize the value of externally developed technology

Locks companies into particular strategies

Creates a strong intellectual property position Differentiates products from those of competitors

87

 

Sources of Innovation

companies disconnect research from development. For example, 3M recently centralized its 12 R&D groups to keep its researchers from becoming too involved in product development at the expense of basic research. 54 Moreover, companies often lose their best researchers and find it difficult to attract new ones if they require researchers to spend too much time on development. Leading researchers want the freedom to select research problems, publish their results, and work on cutting-edge projects. These goals are much more easily accomplished when researchers focus on research than when they conduct development work. Furthermore, transitioning technology from basic research to development is difficult and time consuming. Engineers (who conduct most development work) and scientists (who conduct most basic research) often work in different locations with differentt equipment, design rules, and materials. Therefore, the transition from basic differen research results results to products often involves changing the materials from which a product is made, redesigning it, and changing the process through which it is created. Take, for example, the problems that IBM faces when it wants to transition its semiconductor research to development projects. At IBM, researchers use germanium and gallium arsenide to produce semiconductors in low volume, while the development group uses silicon to produce semiconductors in high volume. Thus, IBM has to redesign its semiconductors to work in silicon when new ne w technologies are transitioned to development from research.55 Finally,, research scientists often have difficulty transferring information about their Finally discoveries to development engineers. Scientists often believe that they are superior to engineers because they “do science,” which leads to low levels of interaction and poor communication between basic researchers and development personnel.56 And engineers, who create new technology products, tend to have very little understanding of  the basic science underlying technology.57 As a result, engineers often need scientific discoveries “translated” for them, which is problematic because scientists and engineers often do not have a common language or format for communication. 58 Several policies encourage productive links between research and development. 1. Giving researchers financial financial incentives to work on projects that are aligned with development goals. Because people respond to financial incentives, these policies help to align research and development. For example, IBM has developed a program to give more funding to researchers who are willing to work more closely with development personnel than to those who will not work closely with them.59 2. Requiring your research laboratories laboratories to use the same equipment and materimaterials as your product development laboratories. For instance, Intel requires its researchers to use the same equipment as development personnel, and to operate that equipment inside of an existing fabrication plant, which minimizes the need for redesign and retesting when discoveries move from research to product development.60 3. Exposing basic researchers researchers to development, development, perhaps for a temporary period period when they first join your company. company. For example, Intel requires all newly hired research scientists to first work in product development, reducing the interpersonal tension between research scientists and development engineers without detracting too much from the scientists’ goal of conducting research. 4. Making fit with company company strategy a criterion criterion in the evaluation of proposals for research only R&D projects related to the capabilities of your firm,funding. as opposedBy to funding everything that shows technical merit or feasibility, 61 then you can better tie together research and development activities.

88

 

Sources of Innovation

GETTING DOWN TO BUSINESS

Investing to Create Opportunities62

Even in a product tech panty hose, companies invest in R&Das tolow come upas with innovations that will increase demand and create competitive advantage over rivals. Some companies are using nanotechnology to design panty hose that have baked-in microcapsules filled with aloe. Because the microcapsules break as the wearer moves, these panty hose help the wearer to combat dry skin. Other companies are adding a derivative of menthol to the yarn that they use to make panty hose. The addition of menthol helps the user feel cooler when she puts on the panty hose, thus making the product more attractive to customers.

investment in R&D in panty hosefunded occurs  bothCorporate in company laboratories laboratorie s and through research at universities. For example, at the industryfunded Textile Protection and Comfort Center at North Carolina State University, researchers investigate new technologies for the textile industry. In fact, the ideas  behind the two two panty panty hose innova innovations tions just describ described— ed— nanocapsules filled with aloe and the addition of menthol to yarn—were developed there. Thus, even in as mundane a product as panty hose, companies conduct R&D to improve their products, and use universities to help them with their innovation efforts.

 Key Points

• R&D is composed of basic research, applied research, and development. • Industry performs more than two-thirds of all R&D in the United States. • Companies invest in R&D to create new technologies that provide the basis for new products and services, to improve their products and differentiate them from those of competitors, to create strong intellectual property positions, and to absorb externally generated ideas. • However, investment in R&D is uncertain; it rarely generates short term financial returns; it locks companies into strategies that are difficult to change; and its  benefits are difficult to appropriate. • Companies need to link research and development to capture the returns to  basic research; however, however, this is not easy because too strong ties divert researchers to short-term problem solving, make employment unappealing to them, and make transitioning technology from basic research to applied development difficult and time consuming.

FO RM S

O F   I NNOVATION

When formulating a technology strategy, you need to consider the different forms that technological innovation can take. While people intuitively think of new technology making possible innovations that take the form of new products or services, those innovations can also take the form of new production processes that allow older products and services to be made in new ways, new raw materials that change the composition of  products or services, new ways of organizing that change the way that existing products are produced and sold, and new markets where the products or services can be offered.63 Table 4 shows some examples of different forms forms that innovations have taken. The creation of the compact an example of innovation, occurred response to the development of adisk ne wistechnology—the new laser—andwhich took the form ofina new product.

89

 

Sources of Innovation

TABLE 4

Relationship Between Technology and Form of Innovation This table shows examples of business ideas that emerged when innovation took different forms in response to technological change. TECHNOLOGICAL CHANGE

EXA XAMP MPLE LE OFA BUSINESS IDEA

FOR ORM M OF TH THE E INNOVATION

Genetic engineering

Synthetic insulin

New product or service

Internet

Online pet food sales

New way of organizing

Refrigeration

Refrigerated ship

New market

Computer

Factory automation

New method of production

Nanotechnology

Smart paints

New raw material

Amazon.com’s effort to sell books on the Internet is an example of an innovation, which occurred in response to the development of a new technology—the Internet— that took the form of a new way of organizing a firm. Amazon.com’s product, the  book, has been around a long time and is exactly the same when sold online as in a  bookstore. The company’s innovation lies in its way of organizing the business as a retail bookstore with no retail outlets. The export of frozen meat by ship to distant locations is an example of an innovation that took the form of a new market. The invention of the refrigerated ship in the 1880s didn’t change the product—the filets were still filets—and the business was still organized the same way way.. The difference was that Argentinean beef could now be sold in Europe as opposed to just in Argentina. The substitution of ceramic composites for metals in the production of motor vehicles is an example of an innovation that took the form of new materials. The product, the car, car, and the market, you and me, are the same whether ceramics are used or not. Moreover, GM is still organized the same way whether ceramics are used in place of metals or not. Even the process of making cars doesn’t change in response to this innovation. Finally, continuous strip production is an example of a new production process used by steel minimills to make an existing product, steel, in a more efficient way. The product, market, and way of organizing the business are the same at Nucor as at U.S. Steel. difference is that Nucor can make steel out. of scrap, which the integrated steelThe mills’ continuous casting process doesn’t allow. allow

Forms of Innovation and Technology Strategy The different forms that innovation can take matter for your company’s technology strategy in two important ways. First, the value that your company can generate and capture from innovating depends on the form that its innovation takes. Sometimes producing old products and services by organizing in new ways, by using new raw materials, or by developing new production processes is more beneficial to a company than developing new products and services. For example, in many industries, the gains from applying nanotechnology lie in the development of new materials or new production processes, not in the development of new products. Moreover, even if value that your company could generate was equal across forms of innovation, the amount of that value it could capture might be different for different forms of innovation. Successful innovation requires companies to develop products and services that competitors will not immediately imitate. The ability to deter imitation is often higher when the innovation takes the form of a new process

90

 

Sources of Innovation

or a new material than when it takes the form of a new product or service. Secrecy is an important barrier to imitation. Because new products and services are sold to customers, their workings cannot easily be kept secret from competitors who can buy a sample and reverse engineer it. However, production processes and input materials can be kept hidden from customers and competitors. Thus, by producing an old product with a new production process, your company will be better able to deter imitation and may capture more of the value from innovating than it will by producing a new product.64 Second, the form that your innovation takes affects your company’s ability to get customers. It is easier to attract customers by targeting a new market with an existing product or service than it is to create a new product or service for an old market. Your Your company’s efforts to innovate only matter to customers if they affect the attributes of the products or services that customers buy buy.. If innovating does not affect those attributes—for example, books are the same whether bought online or at a bookstore—then customers will be more willing to purchase a product than if  innovating changes those attributes. When your innovation changes product attributes, you also need to persuade customers that the new product is better than the old one.

Industry and the Nature of Innovation As a technology strategist, you also need to be aware that the nature of innovation depends on the production process in an industry. Some products, like aluminum, petroleum, and steel, are nonassembled. Lacking components, these products are created through steps or subprocesses that are chemical, thermal, or o r machined. Other products are simple assembled products, made up of linked subsystems, such as stoves, guns, and skis.65 Still other products are assembled systems, composed of distinct subsystems that interact and must be linked together, such as automobiles, televisions, and airplanes. The differences in these production processes affect the nature of innovation in different industries. For instance, technological innovation in nonassembled products is most likely to take the form of improved materials, product performance, 66

processes, production scale. Improving input materialsautomation, is a common way to innovate inor simple assembled products. And mechanization, and standardization, and changes in products’ dimensions, are very common ways to innovate in assembled systems. 67 The locus of innovation also varies across industries. In some industries, like machinery, electrical equipment, and medical instruments, customers are an important locus of innovation. Because customers in these industries have an understanding of their unfulfilled needs, they often innovate to meet those needs, or work closely with their suppliers to develop innovations that would benefit them.68 For example, Singer Sewing Machine Company developed the electronic cash register when it could not get any provider of mechanical cash registers to produce one for it. 69 In other industries, such as food products, lumber, metal working, drugs, soap, and semiconductors, suppliers are an important source of innovation, pushing technological advance to increase demand for their products. For instance, Alcoa and Reynolds Aluminum, two metals companies, developed the two-piece aluminum can and gave the technology to produce it to can manufacturers as a way to increase demand for aluminum.70

91

 

Sources of Innovation

 Key Points

• Innovations can take the form of new methods of production, new raw materials, new ways of organizing, and new markets, as well as new products and services. • Understanding that innovation takes is innovation, important because it affects the value that a firmthe canform generate and capture from as well as the ease of attracting customers. • The nature of innovation varies across industries. DISCUSSION QUESTIONS 1. What are some major sources of innovation? innovation? Why are these things sources of innovation? 2. Should companies invest in research? research? Why or why not? 3. Why is most research conducted conducted by large corporations and not small ones? 4. Why do small firms have a greater greater R&D intensity than large firms? 5. What role does the government play in fostering innovation? Is it necessary to have the government involved to have innovation? Why or why not?

6. Is it a good or a bad thing that universities are engaging in a greater share of R&D in the United States? Why do you take this position? 7. What are some different forms forms that innovation can take? Why do different innovations take different forms?

KEY TERMS Absorptive Capacity: The capability to recognize and use knowledge generated elsewhere. Applied Research: The effort to understand technical or scientific principles with a specific commercial goal in mind. Appropriability: The degree to which the value that results from an investment in the development of an innovation can be captured by the company that

incentive for universities to license the inventions of  their faculty, staff, and students. Complementary Technology: A tech technolog nology y that is used along with a focal technology. technology. Development: The use of technical knowledge to produce something of commercial value. General Purpose Technology: A technology that can  be applied in multiple markets.

made the investment. Basic Research: The effort to understand the technical or scientific principles in a field without particular commercial goals in mind. Bayh-Dole Act: A law that gave universities the rights to federally funded inventions, thus creating a strong

Knowledge Spillovers: edge to another firm. The leakage of a firm’s knowlSubsidy: A payment by the government government that makes up the difference between what customers will pay for a product or service and the cost of producing it.

PUTTING IDEAS

INTO PRACTICE

1. Finding Opportunities for Innovation Many technological innovations occur in response to some kind of change. You can use some of the concepts that were discussed in this chapter to identify an opportunity for technological innovation by identifying the sources of that innovation. Please follow the steps below to do that: Step 1: Identify some changes that you have noticed in (1) technology, (2) demographics, (3) social trends,

(4) regulation, and (5) politics. List the changes  below..  below Changes You Observe

Technology: 1. 2. 3.

92

 

Sources of Innovation

Demographics: 1. 2. 3.

Example: The invention of fuel cells (a technological change) creates the opportunity to make cars that tha t do not have internal combustion combustion engines engines (a new  product).  product ).

Social trends: 1. 2. 3.

1. 2. 3. 4. 5.

Regulation: 1. 2. 3. Politics: 1. 2. 3. Step 2: After you have made this list, think of the following types of technological innovations that

could occur in response to them: (1) new products or services, (2) new markets, (3) new production processes, (4) new raw materials, and (5) new ways of organizing. Please match the source of the opportunity with the type of innovation that the change would lead to.

2. Form of Innovation The purpose of this exercise is to get you thinking about the different forms that innovation can take. Using what you have learned from this chapter, fill in the chart at  bottom of page using us ing examples that were not contained in Table 4. 3. Corporate R&D Pick a company. Go to its Web site or conduct a Google search to find information about its R&D. How much money does the firm spend on R&D? What proportion is spent on basic research and what portion is spent on applied development? Explain how the firm manages the  balance between betwe en “basic” and a nd “applied” “applie d” research. Why does it make the choices that it does? Are the choices “correct”? Why?

NOTES 1. Adapted from Gomez, H. 2006. Tech Tech gadget lets parents watch kids home alone. The Plain Dealer,  January 6: C1, C3. 2. Shane, S. 1996. Explaining variation variation in rates of entrepreneurship in the United States: 1899–1988,  Journal of Management, 22(5): 747–781. 3. Intellectual Property Professional Professional Information Center.. 2006. Perception gap hindering efforts to Center improve patent system, Dudas says. Patent Trademark and Copyright Copyrig ht Journal, 71(1756): February 10, http://ipcenter http://ipcenter.bna.com/pic2/ip. .bna.com/pic2/ip.nsf/id/ nsf/id/ BNAP6LVKRG?.

FOR ORM M OF TH THE E INNOVATION

New product or service New way of organizing New market New method of production New raw material

TECHNOLOGICAL CHANGE THAT MADE IT POSSIBLE

4. http://invento http://inventors.about.com/library/we rs.about.com/library/weekly/ ekly/ aa122999a.htm 5. Vara, V. V. 2007. How firms’ web-filter use is changing. Wall Street Journal, January 16: B3. 6. Eisenmann, T., T., and N. Tempest. 2001. Atheros Communications. Harvard Business School Case, Number 9–802–073. 7. Regalado, A. 2004. Nanotechnology Nanotechnology patents surge as companies vie to stake claim. Wall Street Journal,  June 18: A1, A2. 8. Afuah Afuah,, A. 2003. Innovation Management. New York: Oxford University Press.

BUSINESS IDEA

93

 

Sources of Innovation

9. Sathe, V. V. 2003. Corporate Entrepreneurship. Cambridge, UK: Cambridge University Press. 10. Ramstad, E. 2006. Corning develops a glass without heavy metals. Wall Street Journal, March 22: B2.

34. Klevorick, A., R. Levin, R. Nelson, and S. Winter Winter.. 1995. On the sources of significance of inter-industry differences in technological opportunities, Research Policy, 24: 185–205.

11. buy Clark, D., andekC.for Forelle. 2005. Sun StorageT StorageTek $4.1 billion. WallMicrosystems Street Journal,to  June 3: A3, A6. 12. Holmes, T., T., and J. Schmitz. 2001. A gain from trade: From unproductive to productive entrepreneurship.  Journal of Monetary Economics, 47: 417–446. 13. Abboud, L. 2006. How France became became a leader in offering faster broadband. Wall Street Journal, March 28: B1, B4. 14. Eisenmann and Tempest, Tempest, Atheros Atheros Communications. 15. Sapsford, J. 2005. As Japan’s elderly ranks swell Toyota sees new path to growth. Wall Street Journal, December 21: A1, A12. 16. Metzger Metzger,, R. 2004. Steris making strides in post-terror technology. The Plain Dealer, March 28: G1, G5. 17. McCartney McCartney,, S. 2004. Taxi! Fly me to Cleveland. Wall

35. Ibid. d. , Innovation Management. 36. Ibi Afuah, Afuah 37. Ibi Ibid. d. 38. Science and Engineering Indicators 2006 , http://www.nsf.gov/statistics/seind06/ http://www .nsf.gov/statistics/seind06/c4/tt04–01. c4/tt04–01. htm. 39. Jankowski, J. 1998. R&D: Foundation for Innovation. Innovation. Research Technology Management, 41(2): 14–20. 40. Roberts, E. 2001. Benchmarking Benchmarking global strategic management of technology. Research Technology  Management, March–April: 25–36. 41. Narayanan, V. V. 2001. Managing Technology Technology and Innovation for Competitive Advantage. Upper Saddle River,, NJ: Prentice Hall. River 42. Burgelman, R., C. Christiansen, and S. Wheelwright. Wheelwright. 2004. Strategic Management of Technology and

JournalE. , May 19:Innovation D1, D10. by user communi18. Street Von Hippel, 2001. communities: Learning from open source software. Sloan  Management Review, 42(4): 82–86. 19. Von Hippel, E. 1988. The Sources of Innovation . New York: Oxford University Press. 20. The history of RFID Technology, Technology, http://www.rfidjournal.com/article/articlev http://www .rfidjournal.com/article/articleview/1 iew/1 338/1/129. 21. Afuah Afuah,, Innovation Management. 22. Markoff, J. 2005. A blow to computer science science research. New York Times, April 2: B1–2. 23. Afuah Afuah,, Innovation Management. 24. Thomas, K. 2005. U.S. U.S. funds fuel cell vehicle develdevelopment. The Plain Dealer, March 31: B3. 25. Adams, J. 2005. Industrial R&D laboratories: laboratories:

. NewEntrepreneurship York: McGraw-Hill Irwin. 43. Innovation Sathe,, Corporate Sathe . 44. Afuah Afuah,, Innovation Management. 45. Chesbrough, Intel Labs (A). 46. Hensley Hensley,, S., and R. Winslow Winslow.. 2004. Blood work. Wall Street Journal, April 8: A1, A12. 47. Ibi Ibid. d. 48. Chesbrough, H. 2001. Managing research in Internet time, Harvard Business School Scho ol Teaching Teaching Note , Number 5–601–122. 49. Cohen, W., W., and D. Levinthal. 1990. Absorptive capacity: a new perspective on learning and innovation. Administrative Science Quarterly, 35: 128–152. 50. Song, J., P. P. Almeida, and G. Wu. 2003. Learning-byhiring: When is mobility more likely to facilitate interfirm knowledge transfer? Management Science,

Windows on black boxes? Journal of Technology Technology Transfer, 30(1/2): 129–137. 26. Afuah Afuah,, Innovation Management. 27. http://en.wikiped http://en.wikipedia.org/wiki/Apple_com ia.org/wiki/Apple_computer puter #1975_to_1980:_The_early_years 28. Mansfield, E. 1991. Academic Academic research and industrial innovation. Research Policy, 20: 1–12. 29. Koizumi, K. 2001. 25 years of the AAAS Report: Report: Historical perspectives of R&D in the federal budget, AASS Report XXV , http://www.aaas.org/spp/rd/chap2.htm. 30. Afuah Afuah,, Innovation Management. 31. Adams, Industrial Industrial R&D laboratories. laboratories. 32. Chesbrough, H. 1999. Intel Labs (A): Photolithography strategy in crisis. Harvard Business School Case, Number 9–600–032.

49(4): 351–365. 51. Kim, L. 1997. The dynamics dynamics of Samsung’s technotechnological learning in semiconductors. California  Management Review, 39: 86–100. 52. Song, Almeida, and Wu, Wu, Learning-by-hiring. 53. Argyres, N., and B. Silverman. 2004. R&D, R&D, organization structure, and the development of corporate technological knowledge. Strategic Management  Journal, 25: 929–958. 54. Stevens, T. T. 2004. 3M reinvents its innovation process. Research Technology Management, March–April: 3–5. 55. Chesbrough, Managing research research in Internet Internet time. 56. Chesbrough, Intel Labs (A). 57. Shapero, A. A. 1997. Managing creative professionals. In R. Katz (ed.), The Human Side of Managing

33. Tennenhouse, D. 2004. Intel’s open open collaborative model of industry-university research. Research Technology Management, 47(4): 19–26.

Technological Innovation . New York: Oxford University Press, 307–319. 58. Shapero, Managing creative professionals. professionals.

94

 

Sources of Innovation

59. Chesbrough, Managing research research in Internet time. 60. Chesbrough, Intel Labs (A). 61. Maidique, M., and Hayes, Hayes, R. 1984. The art of high high technology management. Sloan Management Review,

65. Tushman, M., and L. Rosenkopf. 1992. Organizational determinants of technological change: Toward Toward a sociology of technological evolution. Research in Organizational Behavior, 14:

18–31.from Dodes, R. 2005. Weird 62. 25: Adapted Weird science of  pantyhose. Wall Street Journal, November 19–20: P4. 63. Schumpeter Schumpeter,, J. A. 1934. The Theory of Economic Development: An Inquiry into Profits, Capital Credit, Interest, and the Business Cycle. Harvard University Press: Cambridge, MA. 64. Mansfield, E. 1985. How rapidly does technology leak out? Journal of Industrial Economics , 34(2): 217–223.

66. 311–347. Ibid. Ibi d. 67. Klevorick, Levin, Nelson, and Winter, Winter, On the sources of significance of inter-industry differences in technological opportunities. 68. Von Hippel, Innovation by user communities. 69. Afuah Afuah,, Innovation Management. 70. Ibi Ibid. d.

95

 

96

 

Selecting Innovation Projects

From Chapter 5 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 97

 

Selecting Innovation Projects Learning Objectives Real Options: A Vign Vignette ette Introduction Managing Uncertainty Decision-Making Tools Checklists  Analytical Hierarchy Hierarchy Process Process Net Present Value Internal Rate of Return Real Options

Scenario Analysis Decision Trees Portfolio Management Getting Down to Business: Medtronic’s Medtronic’s Use of Portfolio Management Tools Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Identify the different ways that firms manage the uncertainty of innovation, and understand the pros and cons of the different approaches. 2. Describe the different tools for making decisions about innovation, spell out how each of them works, and explain the pros and cons of each. 3. Use the analytical hierarchy process to evaluate innovation projects. 4. Calculate the internal rate of return, payback period, and net present value of innovation projects. 5. Conduct a real options analysis of an innovation project. 6. Explain why real options analysis is often better than net present value calculation for evaluating innovation projects. 7. Use a decision tree to make decisions about an innovation project. 8. Explain why companies need to manage product portfolios, describe the different portfolio management tools that they use, us e, and understand the value of these tools.

98

 

Selecting Innovation Projects

Real Options: A Vignette1  Gemplus International SA, the world’s largest provider of smart cards, is researching wearable, wearable, ultrathin, secure, wireless communication communication devices because the company believes that, through R&D, R&D, it will identify new products and services that will provide it with a competitive advantage. However, However, Gemplus has limited resources and needs to invest its funds wisely, lest it use up its limited resources on poor investments.2 Gemplus’s Gemplus’ s management takes a real-options approach to making R&D decisions. Real options are investments that give a firm the right, but not the obligation, to make an investment. For example, the company has invested in reducing the power requirements for wearable wireless communication devices because lower power requirements could provide the basis for a new family of products based on location services. Moreover, the low-power-technology low-power-technology investments could help the company improve its ability to manage bandwidth if improvements are made in peer-to-peer network architectures. The company is also investing in location technology because such technology would position the company well to compete in service delivery if the market for content-rich service delivery develops.3 By making these initial investments, Gemplus preserves the right to make future investments in the technology should should uncertainty be resolved in a favorable way over time, but does obligate itself to make additional investments. By treating investments as real options, Gemplus minimizes investments in dead ends and preserves flexibility, while while still undertaking the projects necessary to create future products and services. As a result, Gemplus can position itself for the future without straining its financial resources.4

I NTRODUCTION Companies often need to choose between different innovation projects under considerable uncertainty about whether the technological developments are feasible (technical uncertainty), the market needs the innovation (market uncertainty), the resources to put the innovation into place are available (financial uncertainty), and the value from successful innovation can be captured (competitive uncertainty). Doing this successfully requires the use of decision-making tools that are effective for making choices about novel activities when the future is unknown. This chapter explains how to make decisions about innovation projects. The first section examines the strategic approaches that you can take to manage uncertainty. The second section describes several tools that you can use to evaluate innovation projects. The final section discusses the management of an innovation project portfolio, and the tools too ls that support such an activity. activity.

99

 

Selecting Innovation Projects

M ANAGING U NCERTAINTY Innovation is a highly uncertain process. However, researchers have identified four strategic actions that you can take to manage this uncertainty: seeking high returns, minimizing investment, maintaining flexibility, flexibility, and reallocating uncertainty to others. First, you should focus on high return opportunities or you will not be able to generate enough of a financial return from innovation to justify bearing the uncertainty of undertaking it. This means that many innovation opportunities that generate positive financial returns may not be worth pursuing once uncertainty is taken into consideration. Take, for example, a technology with only a one-in-ten chance of success. If you need to generate an average rate of return on your invested capital of 10 percent to satisfy your shareholders, you should only pursue this project if a successful outcome will generate a 100 percent rate of return. Because you will make nothing on nine out of ten projects, a 100 percent rate of  return on the successful project is necessary to get a 10 percent average rate of  return across all projects.5 Second, you should minimize your investment in nonsalvag nonsalvageable eable assets. Minimizing your investment in these assets will help you to manage uncertainty because the cost of   bear  bearing ing uncer uncertain tainty is very much muc h affect affected ed by with the size size of salvage of your potenti your potvalue ential alifloss. loyour ss. Becaus Beinnovation causee you can recoup part oftyyour investment in assets high effort is unsuccessful (but you cannot recoup your investment in nonsalvageable assets), your potential loss is higher with investments in nonsalvageable assets. So what can you do to minimize your investment in assets with low salvage value? You can use standard inputs, like generic machinery, rather than customized inputs. That way, if your innovation effort is unsuccessful, you can sell the generic machinery to another business and salvage some of the value of your investment. In contrast, if you use customized inputs, you will lose the entire value of your investment in them if your innovation effort fails because those assets cannot be sold to others.6 Of course, you will need to use some customized inputs because generic inputs cannot provide the basis of a competitive advantage. Therefore, you need to consider which assets are the source of your business’s competitive advantage when you invest in innovation. Only those inputs that are the source of competitive advantage should be customized. For example, back office accounting software in an accounting firm might make sense to customize, but that same software should not be customized in a medical device firm where it will not create a competitive advantage. You can also minimize your investment by beginning your operations operati ons on a small scale. For example, if you start with a single manufacturing plant and expand only if  you are successful, you can minimize the size of your investment in fixed fixe d assets, and, consequently,, the uncertainty that you need to bear when innovating.7 consequently Finally, you can minimize your investment by turning fixed costs into variable costs. (Variable costs are costs that are dependent on the level of production of your good or service, whereas fixed costs are costs that are independent of your level of  production.) By reducing the amount of your up-front expenditures, you will reduce the size of your loss in the event that your innovation effort is unsuccessful. For example, you might contract with software engineers to write code for you in return a royalty the saleswill of your instead of paying them a salary. As Table 1for shows, this on approach reducproduct reduce e your fixed costs and, consequently, consequently , your  breakeven level of sales. As a result, you will bear less risk in the development of  your software.8

100

 

Selecting Innovation Projects

TABLE 1 Reducing Fixed Costs By turning the fixed cost of writing the software code into a variable cost, the company in this example can reduce its level of initial fixed investment and its breakeven level of sales, decreasing the risk it needs to bear bear.. COST

VERSION 1

VERSION 2

Sofftw So twaare de deve velo lop per erss on sa sala lary ry

Soft So ftwa ware re de deve velo lop per erss ge gett ttin ing g roy oyaalt lty y

Writing software code

$100,000

$0.50/copy sold

Debugging and preparing manual, and packaging

$50,000

$50,000

Distributing software Fixed costs

$0.25/copy $150,000

$0.25/copy $50,000

Variable costs

$0.25/copy

$0.75/copy

Sales price

$10.00/copy

Contribution margin per unit Breakeven level of sales

   

$10.00  $0.25  $9.75



$150,000/$9.75  $15,385



$10.00/copy $10.00  $0.75  $9.25

 

  

$50,000/$9.25  $5,405



Third, you should maintain flexibility so that your company can change direction rapidly if doing so becomes necessary. Because you can’t know in advance the outcome of technical, market, financial, and competitive uncertainty, your ability to change strategic direction when unexpected events occur allows you to avoid a downside loss, and thus to manage uncertainty. Take, for example, the history of computer disk drives. Many manufacturers of  these devices managed the uncertainty of introducing a new product into an unknown market by being flexible about the target market that they were pursuing. When the manufacturers found that the customers that they initially targeted did not want the disk drives that they were offering, they shifted target markets until they found one interested in adopting the drives. 9 Fourth, you should reallocate uncertainty to other parties who are better able or more willing to bear it. For example, DoMoCo, the Japanese phone company, managed the uncertainty of identifying desirable content for its mobile telephone ringtones by creating a competition among content providers. If the content proved to be popular, then the company paid the content provider a royalty, royalty, but if it was not, then DoMoCo bore none of the uncertainty of developing it. 10 Why are other parties willing to bear uncertainty for you? One reason is that they like it. For example, business angels (individuals who use their own money to finance start-ups) often bear some uncertainty for entrepreneurs by investing in their start-ups. They make these investments invest ments because they enjoy the entrepreneurial process and see uncertainty bearing as the price of admission to this activity. Another reason that others will bear your uncertainty is that they are better able to do so than you. Diversified investors, specialists, and companies operating at less than full capacity are all better able to bear uncertainty than others. Diversified investors can bear a lot of uncertainty because they face a low likelihood that their entire investment will be lost in the event of a single adverse outcome. (Diversifying involves investing in a portfolio of uncertain projects with outcomes that are independent of each other.11) For example, venture capital firms can often bear more uncertainty in than the entrepreneurs that they because their simultaneous investment several uncertain ventures with finance uncorrelated outcomes ensures that they will not lose all of their money if one of the start-ups fails. By contrast, each of  the entrepreneurs will lose his or her entire investment if his or her venture fails.

101

 

Selecting Innovation Projects

Specialized investors can better bear uncertainty than unspecialized ones  because  becau se thei theirr spec speciali ializati zation on give givess them info informat rmation ion that reduce reducess the amoun amountt of  uncertainty that they face. For example, a factor (a company that purchases accounts receivable) can collect your bills with greater certainty than you because the factor knows how to collect debt. So specialized investors, like factors, will take on uncertainty that you seek to shed. 12 Companies that are currently engaged in an activity at less than full capacity can  better bear uncertainty than companie companiess that want to start an activity from scratch  because those operating at less than full capacity can undertake the activity in question with a lower potential downside loss. Take, for example, the biotechnology companies with manufacturing plants whose product approvals have been delayed by the FDA. These companies often produce products for other biotechnology companies in their idle plants because the downside loss from an adverse outcome is less for them than for a company that creates a plant from scratch.13  Key Points

• You can manage uncertainty in innovation by seeking high enough returns to  justify the cost of bearing it. • You can manage uncertainty in innovation i nnovation by minimizing the magnitude of your investments in nonsalvageable assets through the use of generic inputs, small scale production, and the transformation of fixed assets into variable ones. • You can manage uncertainty in innovation by maintaining the flexibility to change strategic direction. • You can manage uncertainty in innovation by reallocating it to those parties more willing, or able, to bear it.

D ECISION-M AKING T OOLS As a technology entrepreneur or manager, you should also use decision-making tools for making choices about innovation projects. These The se tools take a wide variety of  forms. Some tools are qualitative, and compare projects on the basis of scales or words, while others are quantitative, and evaluate projects based on the basis of  numerical calculations. Tools can also be comparative, and pit projects against each other, or scoring, and compare projects against standard scales. Different types of decision-making tools have different advantages and disadvantages. Quantitative tools provide a way to get more precise estimates of the contributions of various factors that affect a decision, but can often result in false precision when these factors are not easily quantified. Scoring models incorporate nonfinancial criteria relatively easily, which facilitates thinking about project attributes and strategic factors, but often results in poor decisions when the scores are not precise estimates of  the underlying concepts being evaluated.14 Comparative models allow you to consider different projects in relationship to each other but lead to inaccurate results when you lack information about project quality or need to compare many projects.15

Checklists The checklist is one example of a scoring model. With a checklist, projects are evaluated evaluated on whether or not they meet specific criteria. Checklists are a useful tool whenever the

102

 

Selecting Innovation Projects

FIGURE 1 The Value of  Decision-Making Tools

Decision-making tools help you figure out whether an innovation project is worth doing or not. Source: Washington Post, Post, 2002.

presence or absence of key factors, such as the existing or future capabilities of a firm and its competitors, affects a decision to move forward on a project. 16 For instance, as Table Table 2 shows, a business that is considering the purchase of  software might seek a package that is inexpensive to obtain, support, staff, and use; with a high level of scalability, security, interoperability, and reliability; a long expected life; and easy customization. By using a checklist, the business can determine if these attributes are present or absent in the different di fferent software packages that it is considering.17

Analytical Hierarchy Process Analytic hierarchy process (AHP) is a decision-making tool in which a problem, like the adoption of a new technology, is broken down into a hierarchy of different criteria and choices. When this technique is used, a decision maker starts by identifying the objective, the alternatives, and the criteria on which the decision will be made. For instance, suppose you are considering a new computer system for a factory, and you are trying to choose between four different alternatives. You know that there are three dimensions of the computer system that you care about: compatibility compatibility with the existing system, difficulty in learning how to use it, and cost. But the dimensions are not equally important to you.

103

 

Selecting Innovation Projects

TABLE 2

DIMENSION

A Checklist for Selecting Software Companies often use

PRESENT?

Low cost to purchase Low cost to staff  Low cost to support Low cost to train employees to use

checklists like this one to make decisions about the adoption of  new computer systems.

Easy to customize High level of reliability High level of interoperability with other systems High degree of scalability High level of security Long expected life ch allenges. Cross Source: Adapted from Tuma, D. 2005. Open source software: Opportunities and challenges. http://www.stsc.hill.af.mil/. . ./2005/01/0501tuma.html. Talk , June, http://www.stsc.hill.af.mil/.

The first step in using AHP is to figure out how much you care care about different criteria. To do this, you start by making pair-wise comparisons of the importance of each dimension versus another on a scale of 0 to 10. In this case, suppose that you found that the difficulty difficulty of learning the system is four times more important to you than the compatibility with the existing system; cost is nine times more important than the difficulty of learning the system; and cost is three times more important than compatibility with the existing system.18 You can use these values to create a matrix like l ike the one shown in Table 3. To To do this, you enter the values in the cells that show the relationship between the two dime dimensio nsions. ns.19 You then enter the values for the diagonal (which are always equal to one because each dimension is equally important as itself). Finally, you

TABLE 3 Analytical Hierarchy Process This table shows the analysis for the analytical hierarchy process for the new computer system described in the text. COMP OMPATIBIL ATIBILITY ITY WITH THE EXISTING SYSTEM

DIFFI IFFICUL CULTY TY IN LEARNING THE SYSTEM

COST

Compatibility with the Existing System

1.00

4.00

3.00

Difficulty in Learning the System

0.25

1.00

9.00

Cost

0.33

0.11

1.00

COMP OMPATIBIL ATIBILITY ITY WITH THE EXISTING SYSTEM

DIFFI IFFICUL CULTY TY IN LEARNING THE SYSTEM

COST

1.00/1.58  0.63

 



0.25/1.58  0.16

 



0.33/1.58  0.21

 



Compatibility with the Existing System



Difficulty in Learning the System



Cost



 

4.00/5.11  0.78

 



3.00/13.00  0.23

1.00/5.11  0.20

 



9.00/13.00  0.69

0.11/5.11  0.02

 



1.00/13.00  0.08

104

 

Selecting Innovation Projects

put in the remaining values, which are the inverse of the first set of values that you put in. (For instance, because the difficulty of learning the system is four times as important to you as compatibility with the existing system, the compati bility  bilit y with wi th the t he exist existing ing syste system m is i s one-four o ne-fourth th as a s import i mportant ant as difficu difficulty lty of learn learning ing 20 the system.) Next, you need to calculate the “priority vector,” which is done as follows: First, you calculate the sum of each column. In the example, these calculations are 1.58 (e.g., 1.00  0.25  0.33), 5.11 (e.g., 4.00  1.00  0.11), and 13.00 (e.g., 3.00  9.00  1.00). Second, you divide each entry in the table by its column sum, which is shown in the second panel of Table 3. Third, you take the averages for each row, which aree 0.5 ar 0.555 [e.g.,  (0.63  0.78  0.23)/3], 0.35 [e.g.,  (0.16  0.2  0.69)/3], and 0.10 [e.g.,  (0.21  0.02  0.08)/3]. The row averages make up the “priority “priori ty vector,” or the weighted priority of each attribute.21 The priority vector shows that compatibility with the existing system is the most important dimension, followed by difficulty in learning the system, followed by cost. When making decisions about innovation projects, people are often inconsistent because they do not see the elements of the project as a coherent whole when answering answeri ng questions aboutorchoices betwee between differentmakers dimens dimensions. ions. AHP can be used to see how consistent inconsistent then decision are. If the comparisons are consistent, then the values of the normalized columns in the lower panel of an AHP matrix, such as the one shown in Table Table 3, would all be the same. If the comparisons are not consistent, you can see how inconsistent they are by calculating the priorities scaled against each other, a number that is called the inconsistency ratio.22 If your calculations show that the inconsistency in decision making is very large, it is important for you to understand why before making decisions about the innovation project. Otherwise, you will not be able to produce an internally consistent innovation.23 The reason why AHP is called a hierarchy model is that you can use the values that you calculate to weight decisions at different levels of analysis. analysi s. For instance, you could go through similar calculations for each of the four alternatives that you are considering in the example we have been discussing, and then weight those values  by the priority vector values that you calculated to make your overall decision about the innovation.

Net Present Value Many of the quantitative methods used to evaluate innovative projects are based on the analysis of discounted cash flows. The two most common of these methods are net present value (NPV) calculation, which estimates the value of a project today, given the amount and timing of cash outflows o utflows and inflows and the discount rate; and internal rate of return calculation, which estimates the rate of return on a project, given the level of expenditure and the timing and amount of cash inflows and outflows.24 Table 4 shows an example of a net present value calculation calculation for an innovation project. To calculate the net present value, the cash inflows for each year are divided  by one over the discount rate r ate to the power of the number of years in the future that the inflow occurs. This example shows that, at a 10 percent per year interest rate, a project that generates $1,000,000 per year for five years is worth $3,790,786 in today’s dollars.

105

 

Selecting Innovation Projects

TABLE 4 Present Value of Future Cash Flows This table shows an example of the calculation of the net present value of future cash flows when the discount rate is 10 percent and there are five years of annual cash flows of $1 million per year year.. YEAR

CASH INFLOW

PRESENT VALUE

COMPUTATION

1 2

$1,000,000 $1,000,000

$ 909,091 $ 826,446

PV  CI/(1.10)

3

$1,000,000

$ 751,315

PV  CI/(1.103)

4

$1,000,000

$ 683,013

PV  CI/(1.104)

5

$1,000,000

$ 620,921

Total

$5,000,000

$3,790,786

PV  CI/(1.105) Sum of PV for Years 1–5

PV  CI/(1.102)

The ability to measure the returns from innovation in today’s dollars is important because the investment that you make to generate those returns might have to  be made mad e today tod ay,, and net present p resent value lets you compare co mpare both the t he expendi ex penditures tures and a nd returns in current dollars. For instance, if this project cost $4 million to undertake and the full $4While millionathad be spent sp ent thelook project might not bewould worth undertaking. firsttoglance theimmediately, project might worthwhile—it return $5 million in cash inflows in comparison to the $4 million in cash outflows—the timing of the cash inflows and the time value of money is such that this project would actually yield less in today’s dollars than it would cost to undertake. You can also use net present value calculations to figure out how long it will wil l take you to pay back your initial investment. For instance, if we look at the same example that we have been examining—an up-front investment of $4 million and cash inflows of $1 million per year for five years, with a discount rate of 10 percent—we can figure out the cumulative cash flows by year, which are shown in Table 5. This table shows that at the $4 million investment cost, the project costs are never recouped. However, at a $1.7 million investment cost, the project costs would be recouped by the end of the second year.

Internal Rate of Return Another calculation that you might want to make is for the internal rate of return on a project, which is the discount rate that yields a net present value of zero. The internal rate of return gives you an indication of the financial rate of return that a project

TABLE 5 Payback Period This table shows the payback period for the discounted cash flows shown in Table 4; if the initial investment cost $1.7 million, then the cost of the project would be paid back back in two years. YEAR

DISCOUNTED CASH INFLOW

CUMULATIVE DISCOUNTED CASH INFLOW

1

$909,091

$ 909,091

2 3

$826,446 $751,315

$1,735,537 $2,486,852

4 5

$683,013 $620,921

$3,169,865 $3,790,786

106

 

Selecting Innovation Projects

TABLE 6 Internal Rate of Return This table shows the first iterations of the calculation of the internal rate of return for a project that costs $850,000 to implement and generates $200,000 per year in cash flow flo w over five years. YEAR

0

1

2

3

4

5

TOTAL

Income stream

($850,000)

$200,000

$200,000

$200,000

$200,000

$200,000

$150,000

Present value, at a 5% discount rate

($850,000)

$190,476

$181,406

$172,768

$164,541

$156,705

$ 15,895

Present value, at a 6% discount rate

($850,000)

$188,679

$177,999

$167,924

$158,419

$149,452

($ 7,527)

Present value, at a 5.675% discount rate

($850,000)

$189,260

$179,096

$169,478

$160,377

$151,764

($

26)

could generate. Because you know the rate of return that you can earn on risk-free investments—those in U.S. government securities—you can estimate whether the project provides you with enough of a premium to justify bearing uncertainty to undertake it. The internal rate of return is calculated in an iterative manner. As Table Table 6 shows, you first identify the magnitude and timing of your incoming and outgoing cash flows from the project. Then you discount those cash flows to create a rate of return where the present value of the sum of the initial investment and the annual incoming cash flows is zero. In this example, a 5 percent internal rate of return yields a positive net present value, while a 6 percent internal rate of return yields a negative one. Therefore, the actual internal rate of return must be between 5 percent and 6 percent. By iterating, you will see that 5.675 percent is as close as you can get to a net present value of zero at the three-decimalthree-decimal-point point level. At the time that this text was written, five-year U.S. treasury notes were providing a rate of return of 4.750 percent. Because the innovation project in the example has a five-year internal rate of return of only 5.675 percent, it would make sense as an investment only if it involved slightly more risk than buying the securities of the U.S. government. For most innovation projects, the internal rate of return that is necessary for someone to undertake an investment needs to be much higher. For example, a venture capitalist typically expects an internal rate of return of more than 50 percent to make an investment in a technology start-up at the seed stage—when the entrepreneur has no more than a business plan. That’s because the risk that the investor will lose what he or she invested in the start-up is so much higher than the risk that the investor will lose what he or she lent to the U.S. government. While discounted cash flow analysis is an important tool for making decisions about innovation projects, it faces several important limitations. First, it cannot incorporate many important nonfinancial factors that influence decisions, such as the reaction of competitors, the relationship of one part of the business to another, or organizational learning.25 Second, discounted cash flow calculations do not consider the option value of doing something, like R&D, that could benefit a variety of different projects or make highly profitable future investments possible. 26 Third, discounted cash flow calculations rely on point estimates, which require decision makers to assume (with questionable accuracy) the probability of that outcome occurring when a range of possibilities exist.27 Fourth, the accuracy of discounted

107

 

Selecting Innovation Projects

cash flow calculations depends very much on the accuracy of assumptions about costs, revenues, and time horizons, which is problematic because such assumptions are often very inaccurate for innovation projects. As a result, discounted cash flow calculations favor short-term, high-probability events over long-term, low-probability ones. In fact, discounted cash flow calculations often suggest that firms not make investments in innovation, particularly if the firm already has products with high cash flow. flow.28

Real Options Real-options analysis is a tool that can be used to overcome the limitations of discounted cash flow analysis. As the opening vignette explained, real options provide the right, but not the obligation, to make a future investment. They involve an “option price,” which is the cost of developing a new technology, and an “exercise price,” which is the cost of exploiting the technology. Options are “exercised” when the decision is made to exploit the technology.29 Options have positive value if the expected cash flow from making the investment is less than the value of maintaining the option. Real options are useful for making decisions about investments in innovation for several reasons.InFirst, technology development occursscenario, in an evolutionary fashion thatin is largely staged. the typical technology development initial research results the invention of a new technology, which is followed by product development, manufacturing, and marketing, marketing, in that order. Because Because several things are unknown at each stage of this process, and will only become known if the venture passes through the previous stage successfully, you cannot accurately estimate the value of technology development through all of the stages. Under these circumstances, real options are useful because they limit decision making to information known at the stage at which the decision is being made.30 Second, real options help you to maintain flexibility and avoid committing valuable resources to infeasible alternatives. For example, you might not know if your  biotechnolog  biotech nology y startstart-up up should pursue the human or veteri veterinary nary market for a new drug until you’ve conducted tests on its efficacy on both groups. By using real options to make decisions, you can evaluate the drug’s efficacy on the two groups  before making a decision decisio n about developing a sales force. As a result, you can avoid spending money on a sales force for a market in which the drug doesn’t work. Third, real options permit you to postpone decisions until uncertainty is reduced. 31 Over time, you might gather information about market needs, customer demand, or ways to reduce costs or appropriate returns. 32 Because this additional information helps you to make accurate decisions, delaying your investment until after it has been gathered has value if your investment is irreversible. 33 (For instance, expenditures on the construction of a plant that can do nothing but smelt aluminum are irreversible because the costs cannot be recovered or used for anything else.) By allowing expenditures to be delayed until after information about their value can be gathered, real-options analysis facilitates more accurate decision making. Figure 2 illustrates how real-options real-options analysis can lead to more accurate decisions than discounted cash flow analysis. The figure shows the evaluation of a wireless PDA produ product ct for which which you first need need to invest $12 million million in research. research. This This investment can result in one of three possible outcomes: excellent performance, which has a 30 percent chance of occurring; good performance, which has a 60 percent chance of occurring; or poor performance, which has a 10 percent chance of occurring. After research is completed, the company undertaking the project will need to invest

After research is completed, the company undertaking the project will need to invest

108

 

Selecting Innovation Projects

FIGURE 2 Reall Opt Rea Option ions s Evaluation for a Wireless PDA

Real Options Evaluation for a Wireless PDA

Investment In Research Project

Research Project Outcome

Investment In Product

Commercialization Outcome

Development +$120M

Yes Excellent

0.8 0.2

–$30M

+$30M

No 0.3

+$40M

Yes 0.6

Yes

Good

–$12M

0.3 0.7

–$30M

+$20M

No –$30M

0.1 Yes Poor

0.1 0.9

–$30M

–$120M

No No Valuation Method

NPV

DCA Without Options

–$22.84M

–$10.80M

2006

2007

2008

=

–$12M+

(–$30M/1.12)+

(+$20M/1.122)

=

–$12M+

(–$30M/1.12)+

[(0.30)x[(0.80)x(+$120M)+(0.20)x(+$30M)]+

Most Likely Scenario DCA Without Options Including Uncertainty

(0.60)x[(0.30)x(+$40M)+(0.70)x(+$20M)]+ (0.10)x[(0.10)x(–$30M)+(0.90)x(–$120M)]]  

Real Options Valuation

+4.36M

=

–$12M+

(0.30)x(–$30M/1.12)+

1.122

(0.3)x[(0.80)x(+$120M)+(0.20)x(+$30M)] 1.122

M = Millions

Using real options leads to the calculation of a positive net present value for the wireless PDA project, while not using real options leads to the calculation of a negative net present value. 1996. Applying “options thinking” to R&D valuation. Research Source: Adapted from Faulkner, T. 1996. Technology Management, Management, May–June: 52.

$30 million in product development, which will either yield a viable product or it won’t. If the product is viable, then it can have different potential commercialization outcomes. If the project has an excellent research outcome, the product will have an 80 percent chance chance of earning $120 million and a 20 percent percent chance of earning only $30 million. If the project has a good research outcome, the product product will have a 30 percent chance of earning $40 million and a 70 percent chance of earning $20 million. And if the project has a poor research outcome, the product will have a 10 percent chance of losing $30 million and a 90 percent chance of losing $120 million.

The company has a discount rate of 12 percent per year.

109

 

Selecting Innovation Projects

As the figure shows, the net present value of the most likely scenario—a good research outcome (60 percent chance) and $20 million of commercial value (70 percent chance)—is negative $22.84 million. That is, in net present value terms the company spends $12 million in 2006 and $26.79 million in 2007 ($30 million divided by 1.12) but earns only $15.94 million in net present value terms in 2008 ($20 million divided by [1.12 squared]). Clearly, Clearly, this net present value is not going to motivate the company to undertake the wireless wireless PDA project. If you calculate the net present value to include uncertainty by weighting the stream of payments for probability of the different scenarios occurring, then the net present value of the project is not as bad, but is still negative $10.80 million. The company will spend $12 million in 2006. In addition, in net present value terms, it will spend $26.79 million in 2007 ($30 million divided by 1.12). But the net present value of earnings in 2008 is only $27.98 million because the payments are weighted for the probability of occurring before they are discounted. That is, the expected payment is (0.30)  [(0.80)  ($120M)  (0.20)  ($30M)]  (0.60)  [(0.30)  ($40M)  (0.70)  ($20M)]  (0.10)  [0.10)  ($30M)  (0.90)  ($120M)]. Because the net present value for this approach is still negative, the company again will not undertake the project. if million, we examine the project project will using options, the netispresent valuedifferent is positive But $4.36 and the bereal undertaken. Why the analysis with real options? Although there is no difference in the net present value of the initial research investment when looked at from a real-options perspective as compared to a discounted cash flow perspective, there is a difference in the net present value of the product development expenditure and the commercialization outcome. Because the company knows that it is foolish to persist with its PDA product if the outcome of research is not “excellent,” it makes that investment, if, and only if, the results of the research are excellent. Because the odds that the research results will be excellent are only 30 percent, it only has a 30 percent chance of incurring the expense of the $30 million. As a result, the net present value of that investment is only 30 percent of what it was under the two discounted cash flow scenarios [(0.30)  ($30M/1.12)]. In addition, the net present value of payments in 2008 is also different because the company will only make the product development investment if the outcome of  research is excellent. Thus, there are only two possible scenarios that need to be calculated: the 80 percent chance of making $120 million and the 20 percent chance of making $30 million. As a result, the amount that is discounted to present value for 2008 is (0.80)  ($120M)  (0.20)  ($30M). Of course, this path only has a 30 percent chance of  occurring, so this payment for year 2008 needs to be multiplied by 0.30. However, the potential outcomes on the excellent research research path are so high that the net present value for the project is positive even though there is only a 30 percent chance that the money will be earned. In general, real-options analysis generates a larger value than discounted cash flow analysis if the magnitude of the outcome is large relative to the magnitude of  the investment, when the magnitude of the future outcome is uncertain, when the time horizon is long, and when some information can be gathered in the future that will eliminate uncertainties. 34 Thus, real-options analysis is more appropriate than discounted cash flow analysis for decisions about innovation because the costs, revenues, and time horizons of these projects cannot be measured accurately, 35 there is uncertainty about the value of future payoffs, and project costs are irre irreversib versible. le.36

110

 

Selecting Innovation Projects

FIGURE 3 Scenario Analysis

Scenario analysis generally examines best, worst, and intermediate scenarios. Source: http://www.andertoons.com.

Scenario Analysis analysis, Another toolrepresentation for making decisions about innovation projects is scenarioabout which is the of investments under different assumptions key factors that influence those investments (see Figure 3). Scenario analysis is useful  becausee it helps you to ident  becaus identify ify the source sourcess of uncert uncertainty ainty rathe ratherr than assumi assuming ng 37 them away. By making different assumptions about key variables affecting an innovation project, you can figure what uncertainties the project faces, what factors might lead things to go wrong, and what factors might drive desired outcomes. 38 As a result, you can partial out uncertain from certain outcomes and make decisions about the project more accurately. accurately.39 For example, an entrepreneur writing a  business plan for a busines businesss selling se lling electron electronic ic books bo oks might create different scenarios for the business that considered whether complementary technology in electronic book readers could be developed at different points in time, and whether customers were as willing or less willing to read fiction and nonfiction books in electronic form as compared to paper paper.. Scenario analysis can be made more sophisticated by using Monte Carlo simulation, which is a way to create a probability distribution of outcomes through the use of computer software that experiments with randomly selected values of inputs.40 With Monte Carlo simulation, you can look at the effect of tens of thousands of possible combinations of inputs on outcomes, rather than the effect of only a handful, thus generating more precise results from scenario analysis.

Decision Trees Successful innovation also requires you to make accurate decisions about investments at different stages of the innovation process. The decision tree—a visual representation of decisions and their effects on outcomes, costs, and risks—is a tool that helps you to do that. For example, Figure 4 shows the decision tree for a choice choice between two possible innovation projects. The figure shows that project one will take twice as much R&D expense as project two and will have a 30 percent chance of failing, as compared to no chance for project two. Both projects then require a $1 million investment in manufacturing. The first project has an expected value of only $7.2 million. That is, $4 million  (0.7  $1 million)  0.7  [(0.1  $30 million)  (0.70  $20 million)  (0.20  0)],

111

 

Selecting Innovation Projects

FIGURE 4 A Decision Tree 10% chance

Product 1

70% chance

$1 million invested in manufacturing

$4 million invested in R&D

$20 million sales

20% chance 30% chance

Choose R&D Project

70% chance

$30 million sales

$0 sales Failure

Product 2 $2 million invested in R&D 100% chance

40% chance

$50 million sales

60% chance

$0 sales

$1 million invested in manufacturing

This figure shows a decision tree for the choice between two possible innovation projects.

while the second project has an expected value of $17 million. That is, $2 million  $1 million  [(0.4  $50 million)  (0.6  $0)]. The decision tree clearly shows that the second project is a better choice. Decision trees provide a quantitative evaluation of a choice that is based on the value and probability of outcomes, and that accounts for the influence of staged decision making on risk. However, decision trees are limited by the use of discounted cash flows to calculate the branches on the trees, which makes them subject to all of  the weaknesses of the use of discounted cash flows as a decision-making tool.41



 Key Points

• Companies use quantitative and qualitative, and comparative and scoring, tools to select innovation projects to pursue; the different types of tools have different advantages and disadvantages. • The checklist is a scoring model that evaluates projects on whether or not they meet specific criteria. • The analytic hierarchy process is a decision-making tool in which a problem, like the adoption of a new technology, is broken down into a hierarchy of different criteria and choices. • Many quantitative decision-making tools are based on the analysis of discounted cash flows; the two most common are net present value calculation, which estimates the value of a project today given the amount and timing of 

cash outflows and inflows and the discount rate; and internal rate of return

112

 

Selecting Innovation Projects

calculation, which estimates the rate of return on a project given the level of  expenditure and the timing and amount of cash inflows and outflows. • Real options are an important tool for selecting innovation projects that overcomes the weaknesses of decision making based on discounted cash flow calculations; they give the right, but not the obligation, to make future investments. • Scenario analysis is a quantitative tool for making decisions about innovation projects that examines the effect of different assumptions about key factors. • The decision tree is a visual representation of decisions and their effects on outcomes, costs, and risks; it helps you to make accurate decisions about investments at different stages of the innovation process.

P ORTFOLIO M ANAGEMENT When companies develop multiple products, they often use portfolio management 42

tools to make about innovation. tools help them coordinate the different partsdecisions of the innovation process, setThese up the right order fortothose activities, 43 and determine what resources are needed at different points in the process. They also help decision makers to decide the order in which to pursue projects and the allocation of resources between them, given resource constraints.44 Portfolio management tools include any method that allows you to compare a set of projects against your strategic goals and to allocate scarce product development resources across the different projects. 45 They typically rely on some type of  scoring system that allows you to compare and prioritize development projects along some dimension, such as financial return, strategic fit, or resource demands. 46 Portfolio models take a wide variety of forms, including pie charts and bubble diagrams, and can be based on either quantitative or qualitative analyses. However, project maps are the most important type. As Figure 5 shows, project project maps show the placement placement of projects into into three different categories47—derivative projects (efforts that extend existing projects, like the development of Liquid Tide), platform projects (efforts to create new product families, like the Toyota Camry), and breakthrough projects (efforts based on fundamentally new ideas, like the development of the first digital camera). 48 Project maps help you in several different ways. First, they link product development efforts to strategy strategy..49 Because project mapping helps to allocate resources across different types of projects, you can use it to manage your company’s growth. If you want to accelerate growth, you can allocate more resources to platform and breakthrough projects; whereas if you want to maximize profits from your R&D investments, you can allocate more resources to derivative projects.50 Moreover, project maps help you to formulate the right product development plan, given your business strategy. You You can use project maps to ensure that you have the resources and capabilities for the product development goals that you seek to achieve.51 By analyzing the distribution of your product development projects across types, you can also use project maps to determine the number and mix of projects that you need to achieve your strategic goals. 52 For instance, a project map might reveal gaps in your efforts to develop breakthrough projects and lead you to allocate more human and financial resources to the development of those types of projects,

rather than to derivative ones.

113

 

Selecting Innovation Projects

FIGURE 5 Mapping Product Development Projects

Breakthrough Projects   e   g   n   a    h    C   s   s   e   c   o   r    P    f   o   e   e   r   g   e    D

Platform Projects

Derivative Projects

Degree of Product Change

Mapping development projects is a good waytechnology to ensure astrategists balance between derivative, platform, product and breakthrough project projects s because it helps to visualize the allocation of resources across different types of innovation projects. Source: Adapted from Wheelwright, S., and K. Clark. 2003. Creating project plans to focus product development. Harvard opment.  Harvard Business Review, Review, September: 2–15.

This latter benefit of project maps is particularly important because you need to invest sufficient resources in platform and breakthrough projects for your company to achieve long-term growth.53 However However,, your employees will be much more likely to propose derivative projects than platform or breakthrough projects. Without With out a mechanism to preserve your resources for other projects, derivative projects will use up all of your company’s resources for innovation. Because project maps help you to define your company’s allocation of human and financial resources across different different types of projects, they force you to evaluate project ideas against an overall plan, which helps you to maintain nonderivative projects in your company’s portfolio. Second, project maps help you to avoid overcommitting your organization to innovation.54 Companies often undertake too many innovation projects simultaneously, achieving less innovation than they would have achieved had they undertaken fewer projects. For example, in a study of its own innovation efforts, Exxon Chemical found that it would have had more and better innovation if it had cut in half the total number of projects it undertook. 55 Project maps allow you to measure how many projects you have underway at your company at any point in time, which helps you to avoid overcommitting your company to innovation. Third, project maps help you to sequence your innovation projects to utilize your human resources effectively effectively..56 Research has shown that the productivity of engineers drops dramatically if they have to manage too many projects simultaneously. 57 Project maps provide information about your company’s project allocation, which helps you them to make better decisions hiringyour product development personnel and assigning to projects, therebyabout improving human resource utilization. Fourth, project maps help you to better manage your R&D personnel. Because

engineers learn by doing, they can undertake more complicated innovation projects

114

 

Selecting Innovation Projects

GETTING DOWN TO BUSINESS Medtronic’s Use of Portfolio Management Tools Tools58

Medtronic is a good example of a company that uses portfolio management tools. This Minnesota-based medical device company develops its products off of  common platforms. For example, the company creates its cardiac pacemakers off of a hybrid integrated circuit platform, which allows it to make a variety of  derivative products at low cost. 59 It wasn’t always that way. The company used to develop one product variant at a time, with no derivative products being offered. For instance, the company produced only one version of its very popular Activitrax pacemaker, even though different segments of the market were willing to pay different prices for versions with different features.60 However, the company’s poor performance at product development led its management to require

that all new products be designed so that derivative models could be created from a single platform. 61 The company then used the platform to reach all ends of the market, designing fully featured products to hit the high end of the market and products without some of the features and functionality for the middle and lower end. This approach allowed Medtronic to reach the low end of the market without doing what many companies have to do: cut prices on older models.62 The end result of this effort to create a platform strategy? An increase in the number of derivative products that were developed off of a single platform from 1 to 41.63

as they become more seasoned. As a result, you yo u should first allocate new engineers enginee rs to derivative projects, then move them to platform projects, and finally transfer them to  breakthrough projects. Because project maps provide a layout of your company’ company’ss projects, they facilitate the assignment of engineers to projects appropriate to their level of experience.64  Key Points

• Companies with multiple product lines and multiple products within those lines often use portfolio management tools to align their product development efforts with their technology strategies. • A project map is across a portfolio selection tool thatand helps you to manage the allocation of resources platform, derivative, breakthrough projects. • Project mapping links product development devel opment efforts to firm strategy, helps companies to avoid overcommitment to innovation, facilitates the sequencing of  product development efforts, and improves the management of product development personnel.

DISCUSSION QUESTIONS 1. What strategies can firms adopt to manage uncertainty? What are the pros and cons of these strategies? 2. What are the advantages advantages and disadvantages of difdifferent tools for evaluating innovation i nnovation projects? When might qualitative tools be more appropriate ones,hierarchy and vice versa? When do 3. than Whatquantitative is the analytical hierarch y process? you want to use it? How does it help you to make

4. What problems occur when discounted discounted cash flow analysis is used to make decisions about innovation projects? What decision-making tools should be used if those problems are a re present? Why? 5. How does scenario scenario analysis help you to make decidecisions about innovation projects? Why does scenario analysis help?

decisions about innovation projects?

115

 

Selecting Innovation Projects

KEY TERMS Analytic Hierarchy Process (AHP): A comp comparativ arativee model of project evaluation in which managers create a hierarchy of evaluation criteria. Checklist: The most basic form of scoring model; pro jects are evaluated on on whether or not they meet specific criteria. Comparative Models: Decision-making tools that compare projects against each other other.. Decision Tree: A visual representation of decisions and their effects on outcomes, costs, and risks that people can use to make decisions about ways to achieve a goal. Fixed Costs: The costs that are not dependent on the volume produced. Internal Rate of Return: A calculation of the the payoff of  a project, given the level of expenditure and the timing and amount of cash inflows and outflows. Monte Carlo Simulation: A use of computer computer software to create a probability distribution of financial outcomes based on randomly selected values of inputs. Net Present Value (NPV): An estimate of the financial value of a project today, given the expenditures,

PUTTING IDE DEAS AS

INT IN TO

timing, amount of cash inflows and outflows, and the discount rate. Project Map: A portfolio management management tool that allocates projects into three types: derivative, platform, and breakthrough. Qualitative Methods: Decision-making tools that compare projects on the basis of scales or words. Quantitative Methods: Decision-making tools that select projects based on the basis of numerical calculations. Real Options: A decision-making tool that that is based on the idea that investments give the right, but not the obligation, to make future investments. Scenario Analysis: A decision-making tool tool that reprerepresents investments under different assumptions about key factors that influence those investments. Scoring Models: Decision-making tools that compare projects against standard scales. Variable Costs: The costs that are dependent on the volume produced.

PRACTICE

1. Net Present Value Analysis Your company is thinking of establishing a new semiconductor plant. It will cost you $1 billion to build the plant, all of which will  be incurre incurred d in the first first year year.. From From the the second second thro through ugh the fifth year, the plant will generate $250 million per year in revenue. At the end of the fifth year, the plant

generate $60 million in revenue. Your Your company’s cost of capital is 10 percent. Use the net present and real-options approaches to calculate the discounted cash flows for the project. What does each approach suggest that you do? Why?65 3. Developing Project Maps The purpose of this exer-

will become obsolete andpays will11 nopercent longer generate revenue. Your company to borrowany money.. What is the net present value of this project? money Does the net present value calculation indicate that you should pursue the project or not? Why? 2. Real Options Analysis Assume that you are a manager at a pharmaceutical company where researchers are working on a drug to treat heart disease. The cost of making the drug and the revenues that you will earn from selling an effective drug are uncertain because you don’t know how effective the drug will be or what conditions the drug will treat. You need to conduct $1 million of R&D to develop the drug. If you are successful, you will have to undertake clinical trials that could cost either $25 million or $80 million, with a 50 percent chance of each type of 

cise to develop on a project map forand General Motors. (Forisinformation the company the innovations that it is developing, go to www.gm.com.) Place the following projects on a project map for General Motors, identifying the breakthrough projects, platform projects, and derivative projects: hybrid dieselelectric busses, Chevy Tahoe and GMC Yukon hybrid cars, E85 flex-fuel vehicles, Chevy Volt electric car, Chevy Equinox fuel cell; engine power trains (e.g., LS7-V8 for the Corvette), marine power trains (e.g., Vortec Vortec 8100), Sit-N-Lift power seat; Onstar navigation system, active fuel management, variable valve timing, six-speed transmission, 1.8 Ecotec engine, and displacement on demand. Given the firm’s strategy strategy,, is the allocation of projects across the three types correct? Why or why not? Should the company under-

trial occurring. If you at clinical are two indications forsucceed the drug, each of trials, whichthere has a 50 percent chance of occurring—one which will gen-

take additional projects? If so, what? Should theexistcompany allocate more resources to any of the ing projects? Why or why not? How does this project

erate $40 million revenue, and the other which will

allocation influence what the firm will be able a ble to do?

116

 

Selecting Innovation Projects

NOTES 1. Adapted from Mun, J. 2002. Real Options Analysis: Tools and Techniques for Valuing Strategic Investments and Decisions. New York; Wiley and Sons. 2. Ibi Ibid. d. 3. Ibi Ibid. d. 4. Ibi Ibid. d. 5. Jaffe, A., and J. Lerner. Lerner. 2004. Innovation and Its Discontents. Princeton, NJ: Princeton University Press. 6. Bhide, A., and H. Stevenson. Stevenson. 1992. Attracting Attracting stakeholders. In W. Sahlman and H. Stevenson (eds.), The Entrepreneurial Venture. Boston: Harvard Business School Press, 149–159. 7. Caves, R. 1998. Industrial organization organization and new findings on the turnover and mobility of firms.  Journal of Economic Literature, 36: 1947–1982. 8. Robe Roberts, rts, E. 1991. Entrepreneurs in High Technology. New York: Oxford University Press. 9. Christiansen, C., and J. Bower. Bower. 1996. Customer power, strategic investment, and the failure of  leading firms. Strategic Management Journal, 17: 197–218. 10. Moon, Y. Y. 2004. NTT DoCoMO: Marketing i-mode.  Harvard Business School Note, Number 5–5–3–097. 11. Morris, P., E. Teisberg, Teisberg, and A. Kolbe. 1991. When choosing R&D projects, go with long shots. Research Technology Management, January–Febru January–February: ary: 35–40. 12. Bhide and Stevenson, Attracting stakeholders. 13. Pisano, G. 1995. Nuclean Inc. Harvard Business School Case, Number 9–692–041. 14. Brunner Brunner,, D., L. Fleming, A. MacCormack, and D. Zinner.. Forthcoming. R&D project selection and Zinner portfolio management: A review of the past, a description of the present, and a sketch of the future. In S. Shane (ed.), Handbook of Technology Technology and Innovation Management. Cambridge, UK: Blackwell. 15. Ibi Ibid. d. 16. Schil Schilling, ling, M. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill. 17. Tuma, D. 2005. Open source software: Opportunities and challenges. Cross Talk , June, http://ww http: //www w.stsc .stsc.hill. .hill.af.mil af.mil/. /. . . /2005 /2005/01/0 /01/0501tum 501tuma. a. html. 18. Hallowell, D. Analytical hierarchy hierarchy process (AHP)— Getting oriented, http://www.isixsigma.com/library/conte http://www .isixsigma.com/library/content/c0501 nt/c0501 05a.asp. 19. Ibi Ibid. d. 20. http://msdn.microsoft.com/msdnmag/issues/05/ McCaffrey,, J. The analytical hierarchy process, McCaffrey 06/TestRun.

22. Hallowell, Analytical Analytical hierarchy process (AHP)— Getting oriented. 23. Ibi Ibid. d. 24. Schil Schilling, ling, Strategic Management of Technological Innovation. 25. Schilling, M., and C. Hill. Hill. 1998. Managing the new product development process: Strategic imperatives.  Academy of Management Executive, August: 67–81. 26. Ibi Ibid. d. 27. Narayanan, V. V. 2001. Managing Technology Technology and Innovation for Competitive Advantage. Upper Saddle River,, NJ: Prentice Hall. River 28. Ibi Ibid. d. 29. Hamilton, W. W. 2000. Managing real options. In G. Day and P. Schoemaker (eds.), Wharton on Managing Emerging Technologies. New York: John Wiley. 30. Ibi Ibid. d. 31. Dixit, A., and R. Pindyck. Pindyck. 1995. The options approach to capital investment. Harvard Business Review, May–June: 105–115. 32. McGrath, R. 1997. A real options logic logic for initiating technology positioning investments. Academy of   Management Review, 22(4): 974–996. 33. Dixit and Pindyck, The options approach approach to capital investment. 34. Faulkner, T. T. 1996. Applying “options thinking” to R&D valuation. Research Technology Management, May–June: 50–56. 35. Utterb Utterback, ack, J. 1994.  Mast  Mastering ering the Dynami D ynamics cs of  Innovation. Boston: Harvard Business School Press. 36. Fichman, R. 2004. Real options options and IT platform adoption: Implications for theory and practice. Information Systems Research, 15(2): 132–154. 37. Schwartz, P. P. 1996. The Art of the Long View: Planning  for the Future in an Uncertain World. New York: Currency Doubleday. 38. Ahn, J., and A. Skudlark. 2002. Managing risk in a new telecommunications service development process through a scenario planning approach.  Journal of Information Technology Technology, 17: 103–118. 39. Wright, A. 2000. Scenario planning: A continuous improvement approach to strategy. Total Quality Control, 11(4): 433–438. 40. Brunner Brunner,, Fleming, MacCormack, and Zinner, R&D project selection and portfolio management. 41. Ibi Ibid. d. 42. Rothwell, R. 1994. Industrial innovation: Success, Success, strategies, trends. In M. Dodgson and R. Rothwell The Handbook of Industrial Innovation (eds.), . Aldershot, UK: Edward Elgar, 33–53. 43. Afuah Afuah,, A. 2003. Innovation Management. New York:

21. Ibi Ibid. d.

Oxford University Press.

117

 

Selecting Innovation Projects

44. Brunner Brunner,, Fleming, MacCormack, and Zinner, R&D project selection and portfolio management. 45. MacMillan, I., and R. McGrath. 2002. Crafting R&D project portfolios. Research Technology Management,

53. Schilling and Hill, Managing the new product development process. 54. Burgelman, Christiansen, and Wheelwright, Strategic Management of Technology and Innovation.

45(5): 48–59. 46. Cooper Cooper,, R., S. Edgett, and E. Kleinschmidt. Kleinschmidt. 1997. Portfolio management in new product development: Lessons from the leaders—II. Research Technology Management, 40(6): 43–52. 47. Cooper Cooper,, R., S. Edgett, and E. Kleinschmidt. Kleinschmidt. 2002. Optimizing the stage gate process: What best-practice companies do—II. Research Technology  Management, 45(6): 43–49. 48. Schil Schilling, ling, Strategic Management of Technological Innovation . 49. Schilling and Hill, Managing the new product development process. 50. Burgelman, R., C/ Christiansen, and S. Wheelwright. 2004. Strategic Management of  Technology and Innovation New York: McGraw-

55. Canner Canner,, N., and N. Mass. 2005. Turn R&D upside down. Research Technology Management, 48(2): 17–21. 56. Burgelman, Christiansen, and Wheelwright, Strategic Management of Technology and Innovation. 57. Pelz, D., and F. F. Andrews. Andrews. 1976. Scientists in Organizations. Ann Arbor, MI: University of  Michigan. 58. Adapted from Christiansen, C. 1998. We’ve We’ve got rhythm! Medtronic Corporation’s cardiac pacemaker business. Harvard Business School Teaching Teaching Note, Number 5–698–056. 59. Ibi Ibid. d. 60. Ibi Ibid. d. 61. Ibi Ibid. d. 62. Ibi Ibid. d. 63. Ibi Ibid. d.

Hill/Irwin. 51. Schilling and Hill, Managing the new product development process. 52. Wheelwright, S., and K. Clark. 2003. Creating propro ject plans to focus product development. development. Harvard Business Review, September: 2–15.

64. Wheelwright and Clark, Creating Creating project plans to to focus product development. 65. Dixit and Pindyck, The options approach approach to capital investment.

118

 

Customer Needs Learning Objectives Lead User Method: A Vignett Vignettee Introduction Technology Push Versus Market Pull Understanding Customer Needs  How to Identify Identify Customer Needs Needs The Difficulty of Identifying Customer Needs Significantly Better Benefits Than Existing Products Getting Down to Business: Going Over Like a Wet Tissue  Meeting the Needs Needs of Many Stakeholders Stakeholders

Solutions That Work  Developing Profitable Solutions Pricing Products Correctly Setting a Price Market Segmentation Market Research  Market Research Research Techniques Techniques Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Describe

technology-push and market-pull innovation and

explain how companies should approach market research for each. 2. Define

a customer need and explain when a real customer need

exists. 3. Spell

out why product developers have trouble understanding what customers want.

4. Understand

how product developers determine whether products are economically viable and provide better alternatives a lternatives than competitors’ products.

5. Explain 6. Define

how companies set prices for new products.

market segmentation and explain how companies can segment the market for new technology products and services.

From Chapter 6 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 119

 

Customer Needs

7. Define

market research, describe the different ways to collect market research data, and figure out when each approach to market research should be used. use d.

8. Explain

the advantages and disadvantages of using focus groups, survey methods, ethnography ethnography,, lead user techniques, and iterative methods to identify target markets and customer needs, and identify the market characteristics favorable to each.

9. Identify

when new and established firms have advantages at understanding customer needs and explain why they have advantages at understanding these needs under different circumstances.

Lead User Method: A Vignette1  The Medical-Surgical Division at 3M, a multinational corporation that serves a wide variety of technology markets, was looking to develop new products that would control infection in operating rooms without using antibiotics. The growth of antibiotic-resistant bacteria suggested that products in this segment would offer significant growth potential. Despite believing strongly in customers’ need for products that would control infection without using antibiotics, the managers at 3M did not know what features to incorporate in products to meet that need. The company’s representatives had tried a variety of traditional market research techniques to gather information from customers, but these efforts were not very effective. The hospital staff that the 3M market researchers talked to kept suggesting that the company make incremental changes to its existing products, like like lengthening their their surgical drapes. As a result, the senior management of the Medical-Surgical Division did not feel that traditional market research was providing it with the right information. So the Medical-Surgical Division turned to the lead user method . (We will discuss this method in greater detail later in the chapter chapter.) .) The lead user method is a process for soliciting information from the likely first users of a product. First users are customers who say that they have a need for something that doesn’t yet exist and are experimenting with ways to create a solution to their needs. The lead user method is based on the principal that the best way to develop a new product is to talk to the people who are struggling to find a solution to their needs. If you can understand the needs of lead users, and come up with a solution that meets those needs, then you can develop products that will meet the needs of other users of the product. Many companies swear by the lead user method. Research has shown that it generates faster concept development than more traditional approaches to market research and allows companies to see customer needs for truly new products, rather than just market extensions. So how does the lead user method work? First, the company forms a lead user team, which networks to find lead users in a target market. Once it has identified lead users, the team sets up a workshop in which it seeks to understand their needs. For example, when the Medical-Surg Medical-Surgical ical Division at 3M was trying to understand the needs for the antibiotic-free

120

 

Customer Needs

infection control product, it brought together a MASH unit in Bosnia (which needed products that would work quickly under battlefield conditions) and a veterinary surgeon (whose patients are covered with fur and rarely bathe). Third, the lead user team uses the feedback generated in the workshop to develop new products. In the case of the 3M MedicalSurgical Division, the lead user workshop led the company to develop new draping and antimicrobial products, including a completely new product platform for infection-control devices.2

I NTRODUCTION Efforts to introduce new products often fail because those products don’t satisfy customer needs.3 This is particularly true for new products and services based on new technologies, which are often invented in the pursuit of scientific scienti fic advance, and so are developed in the absence of information about customer needs. 4 This chapter helps you to overcome this problem by offering the tools and techniques that you need to identify customer needs and the features that will satisfy them. The first section compares technology-push and market-pull innovation and explains how companies need to adopt different approaches to understand markets for each. The second section explains why and how companies identify real needs for new products and services, discussing how to identify new products that meet customer needs in a significantly better way than existing products, and satisfy key stakeholders. The third section explains how to price new products correctly. The fourth section describes market segmentation and explains how companies can segment the market for new technology products and services. The fifth section defines market research and identifies the different ways that companies conduct it, outlining the advantages and disadvantages of different approaches under different conditions.

T ECHNOLOGY  P US H  V ERSUS   M ARKET   P UL L Some innovation is driven by customer needs, or descriptions of the benefits that customers want in a product or service.5 For instance, customers might indicate that they have a desire to wear clothes with embedded musical devices, and companies might respond by coming up with a line of clothing with built-in music players. When companies ask customers about their needs and then develop products and services to meet those needs, innovation is called market pull. (Please note that needs are not the same thing as wants wants,, which are a customer’s  beliefs about what products or services will fulfill his or her needs. Customer needs are also different from product attributes because a need is a description of the benefits that customers want in a product or service; whereas an attribute is the way that 6

the product or service gives customers what they need.) Sometimes, people invent new technology in the pursuit of technological advance itself, rather than in response to market needs. Only after that technology is developed do people find the unrecognized “need” for it. 7 The laser represented an 121

 

Customer Needs

FIGURE 1

The First Laser

At the time that the laser was invented, no one knew what the technology could be used for; lasers l asers are now used in CD and DVD players, bar code scanners in grocery stores, eye surgery, surgery, to cut and weld metal, to carry telephone and television signals, and to cut fabric, among a myriad of other things. Source: Corbis/Bettmann.

important technological advance, yet none of the people looking l ooking at the technology at the time had any idea what to do with it. Only later did people realize that you could use lasers to meet a myriad of customer needs (see Figure 1). Because this type of innovation occurs before anyone has figured out a market need that it can meet, it is called technology push.8 Entrepreneurs Entrepreneu rs and managers need to approach market-pull and technology-push innovations differently. differently. For technology-push innovations, recognizing market needs is not the first step in creating a new product because no market exists at the time that the innovation is developed. Instead, the innovator first needs to develop the technology.. Once the technology has been developed, the innovator then needs to find or creogy ate a market by figuring out what benefits it provides and what problems it solves. Some technology-push inventions do not require the creation of a new market,  but rather, rath er, the identific iden tification ation of o f an existing exi sting market m arket for which whi ch the technolo te chnology gy is useus eful. Because the inventors of the technology have come up with the innovation without a particular customer problem in mind, they need to identify a problem that the technology can solve after it has been created. For instance, 3M developed a thin plastic film with microlouvers that worked like a tiny Venetian blind. The company thought it could be used for museum lighting, automatic teller machines, window treatments, and ski goggles, and conducted a variety of focus groups to try to identify the best application for it, without success. It was only when the executive in charge of the project noticed a secretary at 3M blocking the view of her work station from passersby that the application for computer screens was identified.9 Technology-push innovations have several characteristics in common that tend to differentiate them from market-pull innovations: First, they are developed without guidance from customer needs by technologists who are trying to advance knowledge or for their own purposes. Second, they are often based on radical technological changes that open up new markets, rather than on incremental technological changes within existing markets. Third, they often require a long period of 

122

 

Customer Needs

FIGURE 2

Identifying a Real Need

Sometimes people develop products or services for which there is no customer need; if no need ever materializes, those products will be unsuccessful. Source: United Features Syndicate Inc. September 12, 1998.

development during which they are adapted to fit market needs. Fourth, they often face slow customer adoption, given an absence of a problem in the marketplace that the technology can solve, the magnitude of the change that customers experience to adopt the technology, and the need to iterate across market segments s egments to find one that 10 is interested in adopting the product.  Key Points

• Developing a new product in response to customer needs is called market-pull innovation. • When companies first develop a technology and then identify market needs, the process is called technology push. • For some push technologies, innovators need to create a market because no market need yet exists e xists for the technology technolo gy.. • For other push technologies, innovators need to find an existing market that has a need that the technology can solve.

U NDERSTANDING   C USTOMER  N EEDS Unless a company is engaged in technology-push innovation, understanding customer needs is a central part of developing new technology products or services. Why? Because customers don’t purchase product attributes—they purchase the satisfaction of their needs.11 And it is very difficult to satisfy customer needs unless you understand what those needs are.12 Unfortunately, many entrepreneurs and managers do not understand their customers’ needs. Take, for example, the case of Thinking Machines Corporation, a company founded to use massively parallel processing to create supercomputers. Although customers had many needs that Thinking Machines’s computing technology could satisfy, the company’s founders never recognized those needs, concentrating, instead, on developing fastest computer inthat existence. Because customers did not need faster computers the than the alternatives existed, Thinking Machines 13 Corporation failed.

123

 

Customer Needs

How to Identify Customer Needs So how do you know if customers need a product or service that your company is thinking of developing? By answering two basic questions: First, do customers have a problem that no existing product or service solves? Second, is your solution to the problem significantly better than existing alternatives? Customers need your product or service if i f it solves their currently unsolved problems. For example, customers need a noninvasive home test that detects hypertension  because there is currently no way for people to test themselves for hypertension and learn that they should get treatment before the disease causes serious damage. Customers also need your product or service if it offers a better solution to their problems than existing alternatives. For instance, i nstance, a real need exists for Plumpy’nut, a peanut-based spread to treat malnutrition in children. In many developing countries, children suffer from malnutrition. Unfortunately, most nutritional supplements need to be mixed with water; and clean water is not available in many of the places where children are malnourished. Plumpy’nut offers a better solution than other alternatives to the problem of malnutrition in children because, as a spread, it does not need to be mixed with water water..14 The “better” solutions to customer problems that a new product provides do not need to be as profound as that offered by Plumpy’nut. They can be as mundane as reduced complexity, lower risk, greater convenience, enhanced productivity, greater happiness, and higher satisfaction.15 For example, online music downloads benefit customers because they eliminate the need to purchase a whole CD to listen to one song, the requirement that a CD contain music from a single artist, the need to go to a store to buy music, and the limited selection available in a physical outlet. 16 Unfortunately, many new technology products do not offer “better” solutions to customer problems. Take, for example, HP’s TouchSmart TouchSmart PC, an $1,800 computer for the kitchen. The computer allows people to write messages to each other, synch calendars, look up recipes, and order groceries online. However, most customers don’t need an $1,800 computer to write notes to the rest of their family when they can use paper and pencils instead. Moreover, synching calendars isn’t very useful in a product that is not linked to office calendars. Furthermore, few people want to pull out a keyboard in a place with limited counter space where they may not be able to sit 17

down, especially when the alternative is to carry their laptops into the kitchen. So how do you know if customers have unsolved problems or problems that could  be solved in a bette betterr way? The answer is to look to potential customers customers for clues. The  best clue is is a custome customerr complaint, complaint, which indica indicates tes that that a potent potential ial customer customer is unhappy unhappy 18 with the status quo. Take, for example, the case of underwriters at several insurance companies who complain that the software that they use to check the driving records of  new clients is hard to use and inaccurate. The fact that several underwriters complain that they face the same problem indicates that there is a need for a better solution. Another clue to the presence of a customer problem is the expression of an unfulfilled wish. An unfulfilled wish indicates that a customer would do something somethi ng differently only if there were a way to do it. A good example of an unfulfilled wish is the number of people who indicate that they would like to vacation in outer space. The fact that people say that they would like to vacation in space when there is no way to do so at the moment provides evidence of an unmet need for the company that comes up with a way to provide “space vacations.” A third way that you can know if a real need exists is to put yourself in the customers’ shoes and see if i f you have a need for the product p roduct you are thinking of develop-

ing. Many companies do just this. For instance, Procter & Gamble has its executives

124

 

Customer Needs

in charge of disposable diapers meet in a room that looks like a child’s nursery and wear glasses that blur their vision to help them think about diapers from the point of  view of children. Similarly, Meganesuper Co., a Japanese maker of eyeglasses, requires all of its employees to wear glasses to work. This requirement has helped the company to come up with innovations that meet customer needs, such as changes to nose pads that minimize slipping.19

The Difficulty of Identifying Customer Needs While the process of identifying customer needs is straightforward, companies often fail to assess these needs accurately for six reasons: 20 1. Product developers often fail to gather information about customer needs  because they overestimate what they already know. know.21 For example, Nokia has had problems selling its cellular telephones recently because they are much fatter than the phones of its rivals. Nokia does not offer slim mobile phones  because its engineers believed that customers would want the television and video features it packed into its phones more than they cared about slimness, which turns out not to be the case. 22 2. Not all customer needs can be discovered by talking to customers. Customers only express needs that they can think of and often don’t have the imagination to think of beneficial things that don’t yet exist. For instance, before e-mail existed, most people did not know that they would benefit from a system that allowed them to send computer messages to each other while walking around. Now, of course, this is such a necessity to some people that they are referred to as CrackBerry addicts because they cannot live without their BlackBerries. 3. Developers often do not know which customers to ask about their needs. Often, companies do not know the right target markets for the products that they are developing. Because customers are myopic, they cannot provide accurate information about the preferences of customers in other market segments. So if a company does not know the right segments to target, it is unlikely to gather useful information about customer needs. 4. Product to be strong believers in thebeliefs value on of the productswhich that they are developers developing.tend Often, developers project their customers, leads them to incorrectly assess customer needs. For example, the developers of  hybrid vehicles tend to care more about saving the environment than the average customer of automobiles and overweight the importance of that attribute to customers. 5. Companies have structures or processes that make it difficult for them to collect information from customers. For instance, Sun Microsystems recently reorganized its sales force so that its sales people sell multiple products. This restructuring has allowed Sun’s salespeople to gather more information about what customers want in server technology rather than predetermining what they need and trying to sell that to them. 23 6. Needs change over time. As technology advances, it creates needs for new products or products with new features. For example, when the first gigabyte hard drive was introduced in 1991 no one could figure out why anyone would need that amount of 

storage because the only things people were storing were spreadsheets and word processing files. Now, however, terabyte drives may not offer enough memory,

given all of the video and still images that people are trying to store.24

125

 

Customer Needs

Significantly Better Benefits Than Existing Products New products must offer benefits to customers that are significantly greater than those offered by existing alternatives. Prospect theory, an important theory of behavioral decision making, explains why. When people have to choose between two alternatives, they do not perceive both alternatives in the same way. Rather, they pay attention to reference points, such as which alternative is seen as replacing the other. The attributes of the product which came first are seen as things that will be “lost” if they choose the other product, while the attributes of the product which came second are seen as things that will be “gained” if they choose it. If people perceived the same value in products when they are viewed as both a gain and a loss, then perceived value to customers would be indicated by the 45 degree line in Figure 3. However, in reality, reality, people perceive greater loss when outo utcomes are framed as negative (deviation below the 45 degree line) and greater gains when outcomes are seen positive (the deviation above the 45 degree line). Prospect theory research has shown that people value losses loss es at two to three times the level of gains. This means that customers judge the attributes of new products relative to the attributes of current products and require new products to provide  benefits two to three times as large as the benefits that they received from old products before they are willing switch. Take, for example, the efforts of an online grocery delivery service. While this service gives customers the convenience of online shopping, it also requires customers to give up selecting produce or going to the store in search of ideas for dinner.. According to prospect theory, ner theory, customers will not adopt online grocery delivery

FIGURE 3

High value

Prospect Theory Diagram

   l   u  a    t   a  c   n    a    h    t      i  n   a   g    e    o  r   e  m      i  v  e    e   c   r    P  e

Reference point Losses

Gains

   l   a    t  u   c   a     a  n    h    t    s  s    l  o   e    o  r   m   e     i  v   e   r  c    P  e

Value Perceived

Low value

When an outcome is framed as a loss of something that people previously had, they perceive more loss than they actually experience, but when an outcome is framed as a gain of something that they

never had, people perceive more gain than they actually experience.

126

 

Customer Needs

GETTING DOWN TO BUSINESS

Going Over Like a Wet Tissue Tissue

Sometimes, managers or entrepreneurs think that they product, they did not achieve anywhere near that have come up with a new product or service that will result. 27 The company’s management was surmeet customer needs when, really, they have not. Take, prised. The company’s market research showed that for example, the case of Kimberly-Clark and wet tis- 60 percent of adult consumers have used some sort sues. In 2001, Kimberly-Clark introduced a premoist- of moistened cleaning method. Moreover, the marened toilet tissue on a roll called “Cottonelle Fresh ket for disposable wiping products was already over Rollwipes.” The company believed that the product, $4 billion.28 which clips on to a toilet-paper holder holder,, would dramatSo what went wrong? Kimberly-Clark didn’t pay ically increase the sales of toilet paper. Its managers enough attention to the complexity of their customers’ developed elaborate projections of the sales of toilet needs. While many people might benefit from moistpaper,, based on the assumption paper a ssumption that consumers would ened toilet paper, most people don’t want to draw use both dry and wet toilet tissues. 25 attention to that need or their use of moistened toilet Kimberly-Clark made a major investment in the paper. The product, which looked conspicuous in a development and launch of the product, which required  bathroom, made users uncomfortable. uncomfortable. Kimberly-Clark over $100 million of R&D and manufacturing invest- didn’t realize that what customers needed was not just ment, and took $40 million of marketing investment to moistened toilet paper, but secret moistened toilet introduce. Confident in the importance of the the new prod- paper. Moreover, customers didn’t want to pay more uct, the company’s press release at the launch of the for moistened toilet paper than they would pay to wet product called Cottonelle Fresh Rollwipes “the most their own paper. As a result, they were unwilling to significant category innovation since toilet paper first pay the premium that Kimberly-Clark charged for the appeared in roll form in 1890.”26 premoistened paper.29 In short, Kimberly-Clark didn’t While Kimberly-Clark’s management expected do a very good job of assessing customer needs for the to generate $150 million in first year sales from the product.

services unless they perceive the convenience of online shopping to be at least twice as large as the benefits of selecting produce or going to the store in search of  ideas for dinner.30

Meeting the Needs of Many Stakeholders You also need to consider the needs of other stakeholders who influence your customers’ buying decisions when developing a new product or service. As Figure 4 shows, even the decision to purchase a very simple product, like the vial stopper that goes in a syringe, can be affected by a large number of stakeholders. In some cases, other stakeholders have such an influence on customer buying decisions that sales cannot be made if their needs are not met. Take, for example, 3M’s development of a virus-proof surgical gown. While 3M’s customers, the hospitals, believed that this product was great because it offered a tremendous improvement in patient safety over standard surgical gowns, this product was not widely adopted. The virus-proof gown was 10 percent to 15 percent more expensive than other surgical gowns, and managed health-care organizations, which are an important stakeholder in medical product purchases, would not 31

pay the extra price.

127

 

Customer Needs

FIGURE 4

Many Stakeholders Influence a Customer’s Buying Decision FDA

Medical Societies

Hospital Pharmacies

Pharmaceutical Companies

Material Suppliers

Rubber Component Manufacturers

 

Distributors

Content Packages Vial Suppliers

Medical Professionals

Patient

Other Pharmacies

This example shows that a large number of companies are stakeholders whose preferences affect the decision to purchase the stopper that goes in a syringe. Source: Elliot Ross.

Solutions That Work While it might sound obvious that developing new products that work is important to having a successful technology strategy, many companies lose out to competitors  because they cannot develop products that work well. Take, for example, the case of  the social networking site, Friendster. Started in 2002, Friendster was Time Magazine’s invention of the year in 2003 and quickly became the dominant social networking Web site. However, the company suffered a dramatic fall in 2004 because engineering problems made its Web site three times as slow as the Web sites of its major competitors, MySpace and Facebook.32 As a result, Friendster’s customers defected to its competitors.

Developing Profitable Solutions Creating a new product that meets customer needs and provides benefits to them of  at least twice the benefits of existing products is still not enough to succeed at new product development. You also need to develop the new product in a way that generates a profit. Unfortunately, many companies cannot figure out how to develop their new products profitably. Take space tourism as an example. No one has yet figured out how to get people into space at less than the cost that only a few people can afford. As a result, no one has established a space tourism business. To do so would only mean creating a business thatproduct loses money. Developing a profitable is difficult for two reasons. First, you have to

figure out a level of sales at which you can make money money.. If all sales lose money m oney,, then

128

 

Customer Needs

scaling up to produce and sell sel l more units will only o nly create greater losses. However, in most businesses, the profitability of transactions is not the same across all levels of  sales. Businesses that involve economies of scale, have increasing returns, or have large setup costs invariably lose money on initial sales but find that the transactions undertaken at higher volume are profitable. Entrepreneurs and managers often think that if they sell more units they will make money at some point. But that doesn’t always happen. Take, for example, many of the Internet start-ups of the 1990s. The entrepreneurs that founded many of  these companies never could figure out a sales level at which they could make money.. So getting larger just caused them to lose money l ose gobs of money. Second, you need to accurately estimate future costs and prices; incorrect estimates will cause you to fail to make money from developing new products, or fail to develop new products when you could have made money doing so. The problem, of  course, is that it is difficult to figure out what costs will be in the future. While the costs of making and selling many technology products decline as companies get better at making them, predicting this pattern with any kind of accuracy is very difficult. For example, Johnson & Johnson, an early leader in disposable diapers, was driven out of that business because it failed to predict that Procter & Gamble would dramatically reduce the cost of the product. Similarly,, it is difficult to predict what prices Similarly pri ces for products will be in the future. For example, several oil companies have developed ways to turn extra-heavy crude from tar sands into crude oil. However, the cost of this process is approximately $25 per barrel, which makes it worth doing when oil prices are high, but did not make sense when the technology was first developed, and oil sold for $12 per barrel.33  Key Points

• A customer need is a description of the benefits that customers want in a prodproduct or service; it is different from a product attribute, which is a description of  how that need gets satisfied. • To succeed at new product introduction, you yo u need to understand customer needs; many companies have failed because they did not understand their customers’ needs. • A product meets a customer need if it solves an unsolved problem problem or provides a  better solution than existing alternatives to problems that have already been solved. • Customer complaints and unfulfilled wishes are clues to unmet customer needs. • Companies often have trouble identifying customer needs because product developers overestimate their understanding of customer needs, incorrectly project their beliefs about the value of product attributes on customers, do not know which target markets to talk to, have structures and routines that inhibit information gathering, try to gather information only by asking customers for it, and fail to recognize how needs change over time. • Because customers weight losses more highly than gains, new products must provide benefits two to three times better than existing products, or customers not adopt them. • will Companies need to consider the interests of all stakeholders that influence the

customers buying decisions.

129

 

Customer Needs

P RICING  P RODUCTS  C ORRECTLY Whether your innovation is technology push or market pull, correct pricing of your product is anaffects important element ofof a potential successful technology strategy. price of  your product the willingness customers to adopt it, your yoThe ur ability to 34 take customers away from your competitors, and your profit margin. As a technology entrepreneur or manager manager,, you have two basic approaches app roaches to setting the price of your new ne w products: price skimming, in which you set a high price to earn the highest possible profit, and penetration pricing, in which you set a low price to get the highest possible market share. 35 The choice between the two approaches depends on the balance between the benefits and costs of charging a high price. Setting a high price will deter potential customers from adopting the product and will risk losing customers to your competitors. But setting a low price will undermine your profitability profitability.. Moreover, a penetration pricing strategy can be very profitable if you use it to gain customers who can be charged a higher price later. However, However, doing so is risky  because you might mi ght lose money mo ney indefinitely indefi nitely if you yo u can’t raise the th e price of your prodp roduct. The effectiveness of a penetration strategy depends a lot on the kind of business you are in. For businesses in which the value of the product to customers increases with the number of customers already using the product, a penetration strategy can be very effective. For instance, PayPal actually paid its customers to use its online payment system initially because it knew that if customers adopted its service, it would be hard for competitors to dislodge it, and it could raise prices later.

Setting a Price In addition to the topics identified in the previous section, and, of course, the demand for your product or service, you should consider four things when you set your pricing strategy: the t he timing of your market entry, the nature of the market into which you are selling, how products are paid for in your industry, and your cost structure. First, you need to understand how the timing of your company’s entry into the market will affect what customers will pay for your new product. Not only is i s there a competitive aspect to this question—entering after a competitor has launched their product means that your competitor’s price will influence the price for your product—  but also you need to think about the relationship between the price of a product and customer adoption patterns. Some customers will pay more to be the first people to have a product. However, early market entry does not always mean that you should price high and skim the market. If your product is based on increasing returns, you might want to do just the opposite. Your company will benefit greatly from broad initial adoption because the profitability of selling your product will increase with the volume sold. Therefore, setting a low price to encourage adoption is often a more profitable approach in the long run if your product is based on increasing returns. Second, need to consider the market in in which you are the product or service. The you range of prices commonly charged a market is aselling basic factor affecting

prices. Even totally new products are limited in their price range because customers will substitute related products if the price of the new product is not within the

130

 

Customer Needs

expected range. For example, suppose you developed a levitating car that can fly over traffic jams in rush hour traffic. Even though there are no other levitating cars on the market, the price that you set for the car is going to be limited by the price of  helicopters, which would be a substitute for the levitating l evitating car. If you try to price your levitating car much higher than the price of helicopters, customers will shift to the substitute product. Another aspect of your market that influences the price that you can charge for your new products is the structure of the sales channels in the industry. Products often travel through intermediaries before going to the end customers, and those intermediaries want to make a profit on their activities. Consequently, the price that you set needs to consider the profit margins that these intermediaries expect to earn  because the price that your yo ur end e nd customer is willing to pay is limited, and includes i ncludes these profit margins. Third, you need to consider how products are paid for in your industry. In some industries, products are paid for in cash, while in others they are paid for on credit. In some markets, companies routinely discount their prices; whereas, in others, prices are largely fixed. You should consider the effects of hidden costs, discounts, and credit when you set prices for your new product because these factors will affect the actual revenues that you receive from your customers. Credit is a good example of this. If you are going to provide credit to your customers, you need to factor that credit into your calculation of the price for your products. What you will earn is actually the list price that you charge less the cost of the credit that you provide. Fourth, you need to consider your cost structure, including the cost of the materials and labor needed to make your product, as well as your development and marketing costs. For instance, if you are going to provide a great deal of after-sales service, have a lot of sales people, or do a lot of advertising, then you need to price in a way that allows you to pay for these marketing costs. In short, whatever price you set has to allow you to make a profit over the long term. And to figure out if you are going to make a profit, you need to know your costs. One very important aspect of your costs that you need to factor into your pricing is the relationship between fixed fixed and variable costs. As Figure 5 shows, fixed costs, things like your rent, do not vary with the quantity of your product that you make; while variable costs, things like the cost of a assembling a product, are a function function of  the number of units that you produce. Because variable costs depend on how much of your product you make, and the volume of sales is difficult to estimate, you might find it difficult to estimate your company’s per unit portion of fixed costs. Consequently, your company’s pricing will be inaccurate. If you believe that your company will produce more than it actually does, then your per unit price will be too low for your costs and your company will lose l ose money. money. The relationship between fixed costs and variable costs is particularly important in those high-technology businesses that sell products based on information, like digital music. The fixed cost of producing information goods is very high, while the variable cost is almost zero. Therefore, the per unit cost depends almost exclusively on the number of units sold, which makes it is very difficult to forecast costs per unit in these industries. Versioning, fixed fee pricing, and bundling are very effective ways to set prices for information Versioning means offeringsets different versions based of a product of  different qualitygoods. at different prices to different of customers on their

needs. Because competition will drive costs down to variable costs, but fixed costs are high, reducing competition is critical for making profitable the sale of products

131

 

Customer Needs

FIGURE 5

$600

Fixed and Variable Costs

$500 $400       t      s      o       C

$300

$200

$100

Variable Costs Fixed Costs

$0 1

2

3

4

5

6

7

8

9

10

Units Produced

This figure shows the relationship between the number of units produced and fixed and variable costs; variable costs increase with the number of units produced, while fixed costs stay constant.

 based on information. By offering different versions of the product to different customers, companies can segment the market and reduce the competition that drives prices down.36 Providing unlimited use for a fixed fee is also valuable because the variable cost of  providing the product is close to zero. Consequently it costs little to serve customers who exceed the average per unit cost under the fixed-fee arrangement, while allowing extra profit to be captured from those customers who do not reach the average per unit cost. Bundling is the process of offering components to customers for purchase as a group, usually for less than the price of components separately. For instance, software products are often sold in bundles, as is the case when a database spreadsheet and word processor are sold as part of a single package for less than the sum of both of those components alone.37 Bundling is a good pricing strategy, strategy, particularly when a product has a high fixed cost and a low variable cost, because it increases the price that a company can charge on some components, helps it expand into other markets, gives it leverage over suppliers, and deters entry by the providers of the unbundled parts of the product. Consequently, often you can make more money selling one component at a low margin to encourage customers to purchase the other components at a high margin than you can by maximizing the price of each individual component. 38 The effectiveness of bundling depends a great deal on the nature of the industry in which you operate. In addition to the fact that bundling works better in industries that a product basedexist on information, also tends to beexist more effective whensell economies of scale in productionbundling and economies of scope in distribu-

tion because the scale and scope economies generate cost savings, permitting higher profit margins, even when the bundled components are sold at a lower price.

132

 

Customer Needs

Bundling has important implications if you are thinking of entering a new market. When new technologies are introduced, companies in related industries are often better positioned than incumbent firms to take advantage of them because their assets allow them to createthe more valuable product than such incumbent firms can offer. Take, for example, entry of large cablebundles companies, as Cablevision Systems Corporation, into the telephone business. These firms are better positioned to offer VOIP VOIP phone service than the the telephone companies because because their assets in cable, Internet, and wireless services permit them to economically bundle telephone, cable television, Internet, and wireless services, while the phone companies cannot  bundle cable television with these other products. 39 While bundling is a useful pricing strategy, you need to be careful how you use it. Your Your company can violate antitrust laws if it has monopoly power in i n a market and you bundle your products because bundling can be used to exert market power to limit consumer choice. For example, examp le, Microsoft faced antitrust action when it bundled its Web browser with its Microsoft Office suite because the company possessed market power in the office software market. Finally, to determine the right price for a product, you need to understand how the relationship between its different components influences their respective sales. One common relationship between components occurs when one component is purchased once (for example, a cable box) and another component is purchased repeatedly (for example, monthly cable service). For these types of  products, you probably want to price the nonrepeat component low to encourage customers to make an initial purchase into your system. If you can get customers to make the initial purchase, then they will begin to make the repeat purchases that are profitable for you. For instance, with cable service, you might want to charge a low price for an initial setup of the box and make your money on the monthly service fee. Even if you factor all of these things shown in Figure 6 into the price that you set, you have to be ready to dramatically change your pricing structure after you enter the market if your pricing is incorrect. Even giant companies, like Microsoft, have

FIGURE 6

Factors That Influence Pricing

Timing of Entry

Cost Structure

Price

How Products Are Paid For Characteristics of the Market

Four categories of factors influence i nfluence the prices charged for technology products.

133

 

Customer Needs

had to make huge changes in the prices of their products after introduction when the information that they gathered from the market revealed that their pricing structure was wrong. For example, when Microsoft introduced the Xbox, the company quickly found out that to it had get consumers buy priced it.40 the product too high and had to cut the price to $100 to  Key Points

• To launch new products successfully, companies must decide whether to use a skimming or penetration strategy and must set the right price, considering all relevant factors. • Effective pricing depends on many factors, including how the timing of your company’s entry into the market will affect what customers will pay for your new product, your cost structure, the market in which you are selling the product or service, and how products are paid for in your industry. industry. • One very important aspect of your costs that you need to factor into your pricing is the relationship between fixed and variable costs, particularly in those high-technology businesses that sell products based on information. • The high fixed and low variable cost of information goods makes versioning, fixed fee, and bundling good pricing strategies in information-based industries. • Bundling is a good pricing strategy because it increases the price that a company can charge on some components, helps it expand into other markets, gives it leverage over suppliers, and deters entry of the providers of the unbundled parts of the product. • Even if you factor all relevant information into the price that you set, you have to be ready to dramatically change your pricing structure after you introduce a new product if the price that you set turns out to be incorrect.

M ARKET  S EGMENTATION Companies generally face one of three types of markets: homogenous markets, in which all of the customers have the same needs; diffuse markets, in which few customers have the same needs; and clustered markets, in which customers fall into a small number of groups on the basis of their needs. When markets are composed of  customers with clustered demand, market segmentation, the process of dividing a market into groups which have common needs, is often an effective approach to identifying those needs. Market segmentation is particularly helpful when: 1. The technology underlying the product or service is only effective in solving some customer problems and not others. For example, a heart drug named BiDil works on African Americans, but does not work on people of other races, suggesting the value of segmentation in marketing this drug. 41 2. The importance of the solution that the company can provide varies across groups of customers, with some having a much greater need to buy the product

than others. Forcustomers example, when Match.com, anthat Internet dating site,oftargeted Internet savvy it found the ease of use its socialolder, net- less

working Web site made it more attractive than competitor sites to the older segment of the on-line dating market. 42

134

 

Customer Needs

3. The organization lacks the ability to serve all of the market at once, perhaps  because it is a new company that cannot scale up quickly. For instance, a startup software company many not have the resources to develop a version of its

product for from both the and segments thethese market, so will benefit the consumer opportunity toindustrial segment the marketof into twoand groups and go after only one initially. 4. Competitors respond differently to efforts to sell to different groups of customers. For instance, Canon succeeded in getting inroads into the photocopier market by targeting small and medium-sized businesses. Xerox, which was focused on large corporations, did not respond quickly to this attack because it did not care that much about smaller customers.43 In an ideal world, companies would always segment the market by dividing customers into categories based on their needs. However, many needs, such as a desire for high quality, quality, are unobservable. Therefore, to divide the customers into groups that have common needs, companies use “profilers,” which are measurable dimensions, like geographic location, industry, demographics (e.g., age, gender, education), or psychographics (e.g., lifestyle, personality), that are correlated with customer needs. To segment the market, companies identify key profilers and gather information about customer needs. They then divide the market by those profilers and identify the minimum number of profilers necessary to specify the groups of customers that have similar needs within the group, and different needs between the groups. Market segmentation is particularly important to new technology firms. New firms will be more successful if they can avoid drawing the attention, and direct competition, of large, established companies, which they can do if they avoid the latter’s primary customers.44 In addition, they tend to benefit relatively little from economies of scale, given their small size. Segmented markets provide an opportunity for them to enter with small-scale production and focus on underserved niches. Take, for example, the efforts by Nucor, a minimill, to enter the steel industry. Because Nucor initially targeted the segment of the steel market where the profit margins were the slimmest and where only the marginal customers of the integrated steel mills were found, the major steel makers accommodated its entry. Had Nucor gone after the high margin mainstream of the market first, the major steel makers would likely have retaliated and possibly driven it out of business.  Key Points

• When markets are composed of customers with clustered demand, market segmentation, the process of dividing a market into groups that have common needs, is often an effective approach to identifying those needs. • Market segmentation is particularly helpful when the technology underlying the product or service is only effective in solving some customer problems, the importance of the solution that the company can provide varies across groups of customers, the organization lacks the ability to serve all of the market at once, and competitors respond differently to efforts to sell to different groups of  customers. • When needs are unobservable, companies use “profilers,” which are measurable dimensions, like demographics, that are correlated with customer needs. • New firms benefit from market segmentation theycompanies can avoid if drawing the attention, and direct competition, of large, because established they

focus on a niche, and because they tend to benefit relatively little from economies of scale.

135

 

Customer Needs

M ARKET  R ESEARCH Companies often engage in market research, the process of gathering and analyzing information about customer needs, preferences for products, and purchasing decisions. Although market research has important limitations, it is also valuable to companies in five ways:45 • First, it identifies potential customers and the market segments that are most interested in a product or service. • Second, it specifies the product features that customers prefer, and which product developers need to include in new products and services. • Third, it identifies the factors that influence customer purchasing decisions. • Fourth, it provides information about the attractiveness of a market, its size, and its growth. • Fifth, it indicates ways that companies can influence sales, such as cutting prices.

Market Research Techniques

Market research can take a variety of different forms. For example, it can involve the collection of primary data, or it can rely on the analysis of secondary data. It can be qualitative, or it can be quantitative. As a technology strategist, you need to  be aware of the adva advantag ntages es and dis disadva advantag ntages es of the diffe different rent form formss of mark market et research so that you can adopt the right approach when your company develops a new product.

FIGURE 7

The Value of  Market Research

Market research helps you to formulate a technology strategy. strategy. Source: http://www.cartoonstock.com.

136

 

Customer Needs

Newness of the Market 

The effectiveness of many market research techniques rests on three crucial assumptions. First, potential customers understand their own needs and preferences.such While is a pretty fair assumption for products already exist, asthis personal computers or automobiles, it mayand notservices be true that for new-to46 the-world products, which potential customers can’t visualize. Take, for example, Internet shopping. When the Internet was first created, people really did not know how it could be used to meet their shopping needs. As a result, efforts to survey people about their preferences for it were very ineffective. People did not yet understand why they might want to shop online, or even what the concept of  online shopping meant. The second assumption is that companies know who the right customers are for their new products. However, when products are based on new-to-the-world technologies, companies often do not know the right potential customers to talk to  becausee these techn  becaus technolog ologies ies turn out to be most appro appropria priate te for mark markets ets that are 47 different from those that the developers intended. Take the laser, for example. When this technology was first invented, it was not powerful enough to be used for the weapons that the military hoped to use it for. It turned out that the earliest applications of lasers were for surgery. Thus, the developers of the laser initially thought to talk to the wrong set of customers—the U.S. military—about the use for the technology technology..48 The third assumption is that customers can understand the product concept  beforee the new produ  befor product ct has been deve develope loped. d. Howe However, ver, this is ofte often n not the case for new-to-the-world products. Take, for instance, the photocopying machine. When the first photocopier was introduced by Haloid Corporation, the precursor to Xerox, people simply could not easily reproduce existing documents in libraries, universities, offices, and so on. This made it difficult for the developers to explain the product concept to potential customers before they could demonstrate a prototype. Research has shown that using traditional market research techniques to gather information about new-to-the-world products, whether those products are video cassette recorders, xerox machines, e-mail, personal computers, or optical fibers, almost always creates problems. When companies cannot be sure who the right potential customers are, potential customers do not understand their own needs and preferences, and potential users cannot grasp product concepts in advance of their development, then traditional market research techniques just lead to the collection of bad data. For instance, when Corning first developed optical fibers, it went to AT&T, the dominant phone company at the time, and asked about its interest in fiber optic cables. AT&T’s response was that there was no need for this technology in longdistance phone lines, despite the fact that MCI developed its entire phone network on the basis of fiber optics shortly thereafter thereafter.. As a technology entrepreneur or manager, you need to figure out if your products and services are new to the world, like the Apple Newton (the first personal digital assistant),49 or merely a new variation of an existing product, like the latest Intel microprocessor.50 If your products are new to the world, then you need to use market research techniques that are very different from the standard market research techniques that you might have learned about in a market research course. As Table market research on new-to-the-world products, youTable need1 shows, to rely when on a conducting more inductive approach, and you need to get deeply

involved with potential customers to understand the context in which they might use your products.

137

 

Customer Needs

TABLE 1

Techniques for New-to-the-Wor New-to-the-World ld and Less Novel Products This table shows the different approaches to conducting market research for less novel and new-to-the-world products; the latter demand more inductive approaches than the former. ESS

L

OVEL

N

EW TO THE

N

- -

ORLD

-W

Philosophy of Market Research

Deductive analysis

Inductive reasoning

Techniques for Gathering Customer Information

Focus groups, surveys

Lead user method, Delphi technique, ethnography

Examples

New computers, new car models

First Internet auction houses, initial photocopiers

Source: Based on information contained in Barton, D. Commercializing Technology: Technology: Imaginative Understanding of User Needs. Needs. Harvard Business School Note 9-694-102.

When companies do not know who the right potential customers are for their new products, and when they cannot ask customers about their needs and preferences, they often need to engage in an iterative approach to market research, developing an initial version of the product and testing it out on the market to gather the information necessary to further refine it.51 This approach might involve the creation of prototypes with features that would be included in future versions of the product if customers were to respond positively. 52 It might also involve beta testing, or the release of an early version of a product to see how customers react to it before it reaches commercial production.53 By taking what they learn from testing a product with customers, companies can modify their products to meet customer needs. This allows them to launch new-to-the-world products with features that customers want, even though potential customers are unable to communicate their needs or preferences. For example, GE identified the features for its first CT scanner by introducing a crude version on which the company solicited feedback. The feedback enabled GE to improve imaging resolution and speed, and design a product that met customer needs.54 However, the use of an iterative approach to market research is not without risks. When a company introduces an early, and, necessarily flawed, version of its product to obtain customer feedback, it risks its reputation. If the initial version is too flawed and cannot be improved to meet customer needs, then the customers’ view of  the company will suffer suffer.. Because taking an iterative approach to market research has costs and benefits, you need to balance the risk of coming out with a failed product with the cost of  obtaining inaccurate customer feedback. When risk of new product failure is low, but the cost of obtaining inaccurate feedback is high, then you should take an iterative approach to market research. However, when cost of obtaining inaccurate customer feedback is low, but risk of a failed product is high, then you are better off not taking that approach.55 Ethnography. When your potential customers don’t understand their own needs and preferences, and so cannot articulate them, you need to identify customer needs by using market research techniques that involve observation, such as ethnography—the description of a group and its activities based on

observation and participation. 56 By observing potential customers, you can identify needs that customers may not even know they have, let alone are able to

138

 

Customer Needs

articulate. For instance, a product designer might observe a surgeon at work and see the value of a new surgical tool that the surgeon was unaware that he or she needed.57 customers understandoftheir ethnography often helps companiesEven to getwhen a detailed understanding howneeds, customers might use new technology products. For example, when SIRIUS Satellite Radio wanted to figure out how to make a device that would allow the company to gain market share against its larger competitor, XM Satellite Radio, it had a team of ethnographers follow 45 people over a one-month period to see how they listened to music, watched television, and used other electronic devices. The information gathered by the team of ethnographers helped Sirius come up with the S50, a small music playing device that met customer needs much better than the comparable product from XM Radio.58 The use of ethnography is so important to market research in some industries that some companies even have ethnographers on staff whose job is to use the technique to come up with new product ideas. For instance, one Intel ethnographer came up with the idea for the Community PC, which allows millions of farmers in developing countries to access the World Wide Web, after spending two years traveling through the developing world; while another thought of the Centrino wireless technology after observing that Alaskan fishermen needed wireless transmission of  information to adhere to government limits on the amount of different species they were allowed to fish.59 Ethnography is also useful when companies want to enter an industry that is new to them. Take, for example, the case of GE, which wanted to enter ente r the plastics fiber business. Because GE couldn’t just go to potential customers and ask them how to take the market away from the established players, it persuaded several customers to let it visit and tape record conversations between their executives. From this effort, GE learned that its senior executives were wrong in their initial assumptions about what customers wanted. Customers, Custome rs, it turned out, did not care as much about price as about the opportunity to collaborate with suppliers to produce exactly the kind of materials that they needed. So that’s what GE focused on providing.60 The lead user method is a market research technique, developed by Professor Eric Von Hippel of the Sloan School of Management at MIT, which is particularly useful for gathering information about customer needs for new-to-the-world products.61 Standard market research is often limited by the imagination of customers, who do not give much thought to possible new products or services that they might need someday. Consequently, when asked by market researchers about their needs and preferences, most customers request incremental extensions to existing products or services. Lead users are often a better source s ource of information than representative customers  because they are actively seeking new solutions to their problems. As a result, they are a good source of information about the needs of customers for features in new-tothe-world products.62 Moreover Moreover,, lead users are very interested in becoming involved in beta tests and are very likely to be early adopters of any product that solves their problems.63 The lead user method differs from standard market research methods because information is solicited only from people who are so unhappy with existing products that they are trying to develop products to meet their own needs. In addition, lead Lead User Method.

users can help to design new products, rather than just give feedback on them,  because they are involved in product development.64

139

 

Customer Needs

Lead user research is conducted as follows: A company forms a team of between four and six people from marketing and engineering to manage the process. In the first phase, the team identifies the products and markets that they would like to pursue andtoengages in experts. activitiesIntothe learn about them,the such as selects readinga trade journals and talking industry second phase, team specific user trend to focus on and develops a plan for interviewing lead users. In the third phase, the team interviews lead users and gathers information to help understand customer needs and to identify a potential product concept. In the fourth phase, the team conducts a workshop with 10 to 15 lead users to fill in information that could not be gathered in the prior phases. 65 Many companies swear by the lead user method and use it religiously to develop new product concepts. For example, 3M has developed a policy of working with lead users to develop new products and has invested significant resources in the lead user method. As a result, this method has been used to identify a variety of product concepts, concepts, including those for mountain mountain bikes, Gatorade, and sports spo rts bra bras. s. Focus Groups.

When new products are not new to the world, companies

usually who the talk toconcepts are, potential customers generallyknow understand theirright owncustomers needs, and to product are typically easily explained. As a result, companies can use traditional market research techniques to gather information. For instance, they often use focus groups, or discussion sessions of between eight and twelve individuals, who meet for two to three hours to discuss their needs, preferences for different product features, or interest in new products and services. Focus groups are usually led by a trained moderator whose goal is create an atmosphere that motivates the participants to engage in discussion. Focus groups have several advantages as market research tools. First, they have high response rates, reducing the nonresponse bias present with other market research tools, like surveys. People tend to prefer to participate in focus groups than to respond to surveys because focus groups involve the social interaction of a discusd iscussion. Moreover, focus group participants are more likely to answer all of the questions asked of them because of the effect of being in a group setting. Second, the focus group format permits market researchers to ask more openended questions than is the case with many other methods of collecting market research data, like surveys. The use of open-ended questions allows market researchers to gain a greater understanding of customer needs and preferences than is present with closed-ended questions. 66 On the other hand, focus groups have several important limitations. First, they are relatively expensive for the amount of information they provide. Research has shown that two one-on-one interviews elicit approximately the same amount of  information as a single focus group, but cost far less to conduct. 67 Second, the data obtained from focus group respondents are rarely independent of the views of other focus group members. Not only do a small number of  people usually dominate a focus group discussion, leading to an overrepresentation of their views in the information obtained, but also the comments made by one participant influence the subsequent comments made by other participants. As a result, many comments made in focus groups do not represent the participants’ true preferences, but, rather, their interpretations of the preferences of the other

participants.

140

 

Customer Needs

Third, focus groups are hard to control, with the participants taking the discussion in directions that market researchers often do not want to go. As a result, focus group moderators are often confronted with the choice of gathering large amounts of  irrelevant information or Unfortunately, adversely affecting free flowreduce of information vening in the discussion. both the alternatives the valueby of interfocus groups as a market research tool. Fourth, it is often difficult to assemble a focus group that is representative of a larger population. Because focus group participation takes time and effort, many people are unwilling to participate in them. This means that focus groups are often selected samples, making it difficult for market researchers to generalize from focus group data.68 Survey Research. Companies also use survey research, which consists of efforts to ask a sample of people questions about something of interest, such as their preferences for a product or service. For example, you might conduct a survey to find out how product features influence computer purchasing decisions. Companies need to survey a sample, or portion of the population, when they can’t gather market research information from the entire population. For example,

you might want to survey a sample of car buyers about their preferences for product features because the entire population of automobile users is over a billion people. To gather accurate information through the use of surveys, you need to develop questions that people are willing to answer and phrase them in a way that is clear to the respondents. You also need to gather information from a large enough number of people to differentiate statistically between the answers of the different respondents. respondents. Furthermore, you need to identify a sampling frame, or targeted group of respondents, that represents the population about which you seek to gather information. If the subjects in the sampling frame are not representative of the population from which it is drawn (that is, they don't have the same distribution of characteristics), you won’t be able to generalize, or draw inference, to the  broader populati p opulation. on. Groups that differ from the overall population on characteristics that affect customer preferences or purchasing decisions make poor sampling frames for market research. For example, students standing in line at a concert might be a convenient sample to ask about interest in iPods, but they would not be representative of the population of all students. They are very likely to differ from the overall population in terms of their interest in music, particularly live musical events, both of which might be related to their interest in iPods. New and Established Businesses 

Established companies have an advantage over new companies when the market and customer preferences for the innovation are well known. Why? Because these innovations are ones for which the capabilities of established firms tend to be the most beneficial. Large, established firms often develop expertise in market research  based on large sample data collected from surveys and focus focus groups. New firms cannot easily compete with established firms at gathering and analyzing these data  because they have limited resources. Therefore, when the market and customer customer preferences for the innovation are known, established companies have an advantage

over new businesses.

141

  

Customer Needs

TABLE 2

Advantages and Disadvantages of Different Approaches to Market Research This table shows that different approaches to market research have advantages and disadvantages disadvantages that you need to consider when selecting an approach. PPROACH ACH TO MARKET RESEARCH APPRO

ADVANTAGES

DISADVANTAGES

Iterative Ap Approaches

Gets us useful in information fr from customers in a new market

Hurts reputation if product fails

Ethnography

Gets information when customers cannot articulate their needs Gets information from customers very interested in adopting the new product

Takes a lot of time and requires special skills Focuses on the needs of a very select part of the market

Focus Groups

Provides open-ended information from a group with high response rates

Survey Research

Gets information relatively cheaply from a representative sample

Expensive, requires expertise to control, and cannot provide independent data from a representative sample People may be unable or unwilling to answer and might not act in the ways that they say they will

Lead User Method

However, when the demand for the innovation is uncertain (because the market is new or customer preferences are unknown), the large sample market research that established companies do well is not very effective. Instead, what works well are small sample efforts to interact closely with lead customers. (If you do not realize why, why, you should go back and reread the earlier sections of this chapter.) These situations favor new firms because entrepreneurs can often make better and faster decisions on the basis of smaller amounts of information than established firms, which have to adhere to established decision-making rules and norms. Therefore, new firms are often better than established firms at gathering information about customer needs in truly new markets.  Key Points

• Market research involves the collection and analysis of information about customer needs, preferences for products, and purchasing decisions. • It consists of both primary research (collecting data from respondents) and secondary research (analyzing data collected by others), and both qualitative and quantitative analysis. • Appropriate market research techniques for new-to-the-world products are  based on induction and intuition; while appropriate market research techniques for less novel products are based on deduction and analysis. • When products are new to the world, companies are often better off using an iterative approach to market research, even though such an approach risks their reputations if they can’t develop those products successfully. • Ethnography is a technique that involves observing customers to learn their needs and preferences, and is useful when customers cannot articulate those needs and preferences. • The lead user method is a way to develop truly new products or services by gathering information from people who are so unhappy with existing products that they are trying to come up with new solutions.

• Focus groups are sessions in which eight to twelve people come together to discuss their needs or preferences for new products or services; they are valuable

142

 

Customer Needs

for gathering fine-grained information from customers about existing product categories but are not very useful for drawing inference to a larger population, particularly for new-to-the-world products. • Survey research asks on respondents questions about their needs, the andselection preferences; its accuracy depends the existence of the product category category, of a sampling frame that is representative of the population of interest, a large enough sample size to draw inference to the population with confidence, and an effective method of data collection.

DISCUSSION QUESTIONS 1. When should you be concerned with with customer needs in the process of product development, and when should you not focus on these needs? Why? 2. Why do many companies fail to develop products products that meet customer needs? 3. Why do companies companies fail with products that do meet

7. What market research techniques techniques do you think are most effective and least effective? Are there conditions under which some market research techniques work better than others? If so, what are those conditions, and why do they influence the effectiveness of  market research techniques?

customer needs? 4. Why don’t customers adopt new new products that are  just as good as competitors’ competitors’ products? products? 5. How should companies companies price their new products? products? What factors should influence their pricing decisions? 6. How should companies companies segment the market? What factors are most important to consider in segmenting a market?

8. What are the advantages advantages and disadvantages of the the lead user research methodology? 9. Can companies be too too market-oriented? Why Why or why not? What are the advantages and disadvantages of being market-oriented?

KEY TERMS Beta Testing: The release of early versions of a product to see how customers react to the product before it

Market Segmentation: The process of dividing a market into groups that have common within group

reaches commercial production. Bundling: The process of offering components to customers for purchase as a group, usually for less than the price of the components separately. Customer Need: A description of the benefits benefits that customers want in a product or service. Ethnography: The description of a group and its activities based on observation and participation. Focus Group: A meeting of people under under the direction of a trained moderator to discuss needs or preferences for products. Lead User Method: A market research technique that that gathers data from customers who would likely be the first to adopt the product or service. Market Pull: The situation in which companies ask customers about their needs and then develop prod-

needs and different between group needs. Penetration Pricing: A pricing strategy in which a company sets a low price to get the highest possible market share. Price Skimming: A pricing strategy in which which companies set a price to earn the highest possible profit. Product Attribute: The way that a customer need is satisfied by a product or service. Representative: Having the same distribution of characteristics as the population from which a sample is drawn. Sample: A portion of the population population contacted to gather market research data. Sampling Frame: A targeted group of of respondents, which represents the population to which one seeks to generalize. Survey Research: Research that asks a sample of

ucts and services to meet those needs. Market Research: The collection and analysis of infor-

respondents questions about their needs or preferences. Technology Push: The situation that occurs when

mation about customer needs, preferences for products, and purchasing decisions.

companies develop a technology and then find a need in the marketplace that the technology can satisfy.

143

 

Customer Needs

PUTTING IDEAS

INTO PRACTICE

1. Is There a Real Need? Identify three technology products that were introduced in the couple years. Identify the customer need thatpast made the of  introduction of those products worthwhile. Then think about how badly customers needed each of  these products. Which one did customers need the most? The second most? The least? Explain why you rank these products the way that you do. Now think about the attributes of these products. How well does each of them meet customers’ needs? Rank order the three products from meets-needs-themost to meets-needs-the-least. Explain why you rank the three products in this order? Is there a relationship between the ranking that you gave the products on the magnitude of  customer needs and how well the products meet customer needs? Why or why not? Role of Stakeholders Identify a technology 2. The product or service with which you are familiar.

Make a list of all al l of the stakeholders that influence the decision of the end user. eachbuying of these stakeholders affects theExplain buyingwhy decision. Make a chart (similar to the one in Figure 4) that shows how the different stakeholders influence the end customer’s buying decision. 3. Approach to Market Research Think of three unfulfilled customer needs that a technology product or service could fill. For each of these needs, describe in one paragraph a product or service that would meet that need. Now assume that you need to gather information from customers about their needs, the attributes that the products would need to meet those needs, and how well the product you described would do this. For each of the three cases, identify the market research technique that you  believe would be the the best for gathering this inforinformation. Explain why you would recommend that approach.

NOTES 1. Adapted from Henderson, Henderson, C. Finding, examining lead users push 3M to leading edge of innovation, www.refresher.com/!leadusers. 2. Anonymous. 2001. The innovation engine. 3M Stemwinder, March 20–April 9. 3. Cooper Cooper,, R. 1979. The dimensions of industrial new product success and failure. Journal of Marketing, 43: 93–103. 4. Burgelman, R., C. Christiansen, and S. Wheelwright. Wheelwright. 2004. Strategic Management of Technology and Innovation . New York: McGraw-Hill/Irwin. 5. Urban, G., and J. Hauser. Hauser. 1993. Design and Marketing of  New Products. Upper Saddle River River,, NJ: Prentice Hall. 6. Bayus, B. Forthcoming. Forthcoming. Understanding customer customer needs. Blackwell Handbook of o f Technology Technology and Innovation Management. Oxford: Blackwell. 7. Mokyr Mokyr,, J. 1990. The Lever of Riches . New York: Oxford University Press. 8. Ettli Ettlie, e, J. 2000. Managing Technological Technological Innovation . New York: John Wiley Wiley.. 9. Bartlett, C., and A. Mohammed. Mohammed. 1999. 3M optical systems: Managing corporate entrepreneurship.  Harvard Business School Case, Number 9–395–017. 10. Markides, C., and P. P. Geroski. 2005. Fast Second: How Smart Companies Bypass Radical Innovation to Enter and Dominate New Markets. San Francisco: Jossey-Bass.

11. Wind, J., and V. V. Mahajan. 1997. Issues and opportu-

12. Zirger Zirger,, B., and M. Maidique. Maidique. 1990. A model of new product development: An empirical test.  Management Science, 36: 867–883. 13. Mullin Mullins, s, J. 2003. The New Business Road R oad Test. London: Financial Times Prentice Hall. 14. Thurow Thurow,, R. 2005. In battling hunger, a new advance: peanut-butter paste. Wall Street Journal, April 13: A1, A14. 15. Kim, W., W., and R. Mauborgne. 2000. Knowing Knowing a winning business idea when you see one.  Harvard Business Review, September-October: 129–137. 16. Moon, Y. Y. 2006. Online music distribution in a postNapster world. Harvard Business School Teaching Teaching Note, Number 5–506–058. 17. Boehret, K. K. 2007. A new push to put a PC in the kitchen. Wall Street Journal, January 24: D4. 18. Cooper Cooper,, R., S. Edgett, and E. Kleinschmidt. Kleinschmidt. 2002. Optimizing the state-gate process: What bestpractice companies do—I. Research Technology  Management, 45(5): 21–27. 19. Dvorak, P. P. 2007. Seeing through buyers’ eyes. Wall Street Journal, January 29: B4. 20. Stein, E., and M. Iansiti. 1995. Understanding user needs. Harvard Business School Note, Number 9–695–051. 21. Gourville, J. 2005. Why developers developers don’t under-

stand why consumers don’t buy. Harvard Business

nities in new product development: An introduction to the special issue.  Journal of Marketing Research, 34(1): 1–12.

School Note, Number 9 504 068. 22. Bryan-Low Bryan-Low,, C. 2006. Nokia joins slimming trend. Wall Street Journal, November 28: B2.

144

 

Customer Needs

23. Lawton, C. 2006. Sun Microsystems remakes its sales force. Wall Street Journal, December 26: B3. 24. Gomes, L. 2007. As disk drives reach new milestone, milestone, flash gains new currency. Wall Street Journal, January 17: B1. 25. Nelson, E. The tissue that tanked, http://www.wsjclassroomedition.com/archive/02s ep/MKTG.htm. 26. Ibi Ibid. d. 27. Bayus, Understanding customer needs. 28. Barblova, I. Uncertain Uncertain fortunes for wet tissue, http://www.paperloop.cpm/db_area/archive/tw_ mag/2003/0307/exit_issues.html. 29. Ibi Ibid. d. 30. Gourville, Why developers developers don’t understand why why consumers don’t buy. 31. Thomke, S. 2002. Innovation Innovation at 3M (A), Harvard Business School Case, Number 9–699–012. 32. Piskorski, M., and C. Knoop. Knoop. 2006. Friendster (A).  Harvard Business School Case, Number 9–707–409. 33. Gold, R. As oil prices surge, oil giants giants turn sludge into gold. Wall Street Journal, March 27: A1, A15. 34. Jedidi, K., and J. Zhang. 2002. Augmenting Augmenting conjoint analysis to estimate consumer reservation price.  Management Science, 48(10): 1350–1368. 35. Schil Schilling, ling, N. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill. 36. Shapiro, C., and H. Varian. 1998. Versioning Versioning the smart way to sell information. Harvard Business Review, November–December. 37. Bakos, Y., Y., and E. Brynjolfsson. 1999. Bundling information goods: pricing, profits and efficiency, efficiency,  Management Science, 45 (12): 1613–1630. 38. Ibi Ibid. d. 39. Brown, K., and A. Latour. 2004. AT&T AT&T will offer Internet phone calls in selected markets. Wall Street  Journal, March 30: B1, B2. 40. Lohr Lohr,, S. 2007. Preaching from the Ballmer pulpit. New York Times, January 28: Section 3, 1, 8–9. 41. Saul, Stephanie. 2005. FDA approves heart drug for African-Americans. New York Times, June 24, http://www.nytimes.com/2005/06/24/h http://www .nytimes.com/2005/06/24/health/24d ealth/24d rugs.html?ex 1172984400&en 51190b4fb32616a3&ei 5070. 42. Silver, S. 2007. How Match.com found love among  boomers. Wall Street Journal, January 27–28: A1, A7. 43. Markides and Geroski, Fast Second. 44. Romanelli, E. 1989. Environments Environments and strategies of  of  organization start-up: Effects on early survival.  Administrative Science Quarterly, 34: 369–387. 45. Cooper Cooper,, R., and E. Kleinschmidt. 1987. New products: What separates winners from losers? Journal of  Product Innovation Management, 4: 169–184. 





47. Friar, J., and R. Balachandra. Balachandra. 1999. Spotting the customer for emerging technologies. Research Technology  Management, 42(4): 37–43. 48. http://www http://www.lasers.org.uk .lasers.org.uk /laser_welding/  briefhistory.htm  briefhistory .htm 49. Barton, Commercializing technology. technology. 50. Sathe, V. V. 2003. Corporate Entrepreneurship. Cambridge, UK: Cambridge University Press. 51. Afuah Afuah,, A. 2003. Innovation Management. New York: Oxford University Press. 52. Brown, S., and K. Eisenhardt. Eisenhardt. 1997. The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations. Admini  Administrati strative ve Science Scie nce Quarterly Qua rterly , 42(1): 1–34. 53. Schil Schilling, ling, Strategic Management of Technological Innovation. 54. Lynn, G., G., J. Morone, and A. Paulson. 1996. Marketing and discontinuous innovation: The probe and learn process. California Management Review, 38(3): 8–27. 55. Krubasik, E. 1988. Customize your your product developdevelopment. Harvard Business Review, November– December, 4–9. 56. Bayus, Understanding customer needs. 57. Leonard, D., and J. Rayport. Rayport. 1997. Spark innovation through empathic design. Harvard Business Review, November–December: November–Decem ber: 102–113. 58. The Science of Desire. Business Week , June 5, 2006, http://images.businessweek.com/ss/06/05/ethno graphy/index_01.htm. 59. Ibi Ibid. d. 60. Ibi Ibid. d. 61. Sathe Sathe,, Corporate Entrepreneurship. 62. Brown and Eisenhardt, Eisenhardt, The art of continuous change. 63. Day Day,, G. 2000. Assessing future markets for new technologies. In G. Day and a nd P. P. Schoemaker (eds.) Wharton on Managing Emerging Technologies. New York: John Wiley Wiley.. 64. Lilien, G., P. P. Morrison, K. Searls, M. Sonnack, and E. von Hippel. 2002. Performance assessment of  the lead user idea-generation process for new product development. Managemen  Managementt Science , 48(8): 1042–1059. 65. Thomke, S., and A. Nimgoade. 1998. Note on lead user research. Harvard Business School Note, Number 9-699-014. 66. Allen Allen,, K. 2003. Bringing New Technology to Market . Upper Saddle River, NJ: Prentice Hall. 67. Griffin, A., and J. Hauser. Hauser. 1993. The voice of the customer. Marketing Science, 12(1): 1–27. 68. Mullin Mullins, s, J. 2003. New Business Road R oad Test. London:

46. Barton, D. 1994. Commercializing Commercializing technology: Imaginative understanding of user needs. Harvard Business School Note, Number 9–694–102.

Financial Times Prentice Hall.

145

 

146

 

Product Development 

Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. From Chapter 7 of Technology Strategy for Managers and Entrepreneurs All rights reserved. 147

 

Product Development  Learning Objectives Product Development: Development: A Vigne Vignette tte Introduction Product Cycle Time Concurrent Development

Product Development Tools Web-Based Tools  House of Quality Quality Prototyping Getting Down to Business: Prototyping at IDEO

Modularity and Product Platforms Product Platforms Identifying the Right Product Features Conjoint Analysis Kano Method Perceptual Mapping Concept Testing

Stage Gates Design for Manufacturing Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Explain

how to identify and prioritize customer features for new products.

2. Produce

a perceptual map, conduct a concept test, do a conjoint analysis, and use the Kano method.

3. Spell

out the pros and cons of reducing product cycle time, and describe how companies do it.

4. Explain

how to use prototyping, house of quality, design for manufacturing, and Web-based tools to improve the product developfacturing, ment process.

5. Describe

the stage gate process and explain how it helps to make decisions about innovation.

6. Define concurrent development and

explain the advantages and

disadvantages of engaging in it. , 7. Define modularity 

and explain the advantages and disadvantages of developing modular products.

148

 

Product Development

Product Development: A Vignette In 1993, Apple Computer introduced the first personal digital assistant (PDA), the Apple Newton, after spending $200 million on the project.1 Although the Newton had a number of innovative features, including spreadsheet software and printing capability, it was a failure (see Figure 1). The product cost as much as a desktop computer, ran very hot, and was heavy. 2 Moreover, customers did not want many of the features of the Newton, including a hard drive and infrared long distance networking capability. In contrast, Jeff Hawkins created the Palm Pilot, a handheld organizer with far fewer features than the Newton: a calendar, address book, to-do list, memo pad, and computer connectivity.3 The Palm Pilot was small enough to be placed in the pocket of a shirt, was very light, and cost considerably less than the Newton. The product was a huge success, and, within a few years, Palm took control of more than two-thirds of the market.4 Why was the Palm Pilot a success when the Apple Newton was a failure? Part of the answer lies in the development of supporting technologies. The Newton predated the Palm by several years, and many of the supporting technologies that have made PDAs so popular had not yet been developed at the time the Newton was introduced. But part of the answer also lies in how the two companies conducted product development. Hawkins discovered that customers did not want handheld computers to replace their desktop computers; they wanted handheld computers to replace their calendars. So, unlike Apple Computer, Hawkins created a product that met customer needs without including unnecessary features that undermined the appeal of the product, and raised its costs. 5 To understand how he did that, and how you can too, read on. Effective product development is the subject of this chapter.

FIGURE 1

The Problems with the Apple Newton

The Apple Newton was not very successful, in part because it used novel handwriting recognition software that didn’t work very well. Source: http://images /images.ucomi .ucomics.com cs.com/ /comic comics/ s/db/1993/db930827.gif  db/1993/db930827.gif 

149

 

Product Development

I NTRODUCTION This chapter discusses product development, or the creation of new products by companies, and its role in technology strategy. Although product development involves a variety of activities, from making improvements to existing products, to developing new product platforms, to creating new-to-the-world products, all product development efforts have one thing in common: They influence the way in which firms are organized and compete. Therefore, you need to think about product development in the context of your company’s technology strategy. strategy. Take, for example, the efforts of two companies to develop microchips for cellular telephones. One competitor, Texas Instruments, has concentrated on meeting a wide range of design specifications; whereas another competitor, Silicon Labs, has focused on keeping costs down. These different strategies have led the two companies to develop very different products. Texas Texas Instruments has developed a product that places transmission activities on one chip and the power function on another chip; while Silicon Labs, has developed a chip that puts all of these functions together in one place.6 This chapter is organized as follows: The first section discusses why reducing product development time is an important issue for companies and describes how to do it. The second section looks at concurrent product development and discusses the advantages and disadvantages of using it. The third section examines modular product development, explaining why companies use it, and what its limitations are. The fourth section describes how to identify the right features for your products and discusses several tools that help you to do that: perceptual mapping, concept testing, conjoint analysis, and the Kano method. The fifth section examines several important product development management tools, including prototyping, the house of quality, design for manufacturing, and Web-based search tools, and explains how you can use them to improve your product development efforts.

P RODUCT   C YCLE  T IM E Product cycle time is a measure of how long it takes companies to develop new products or services. While research shows that the length of the product cycle depends on the type of product (with cycle time being longer for more complex and more novel products)7 and several industry characteristics, companies in a wide variety of industries are reducing product cycle time across a wide variety of new product types.8 Therefore, as a technology manager or entrepreneur, you need to understand how to reduce product cycle time. But, before we turn to explaining how you can do this, you first need to understand why reducing product cycle time is such an important aspect of technology strategy today today.. Reduced product cycle time helps companies in four ways:

• First, it increases the amount of time that your products can be on the market  before becoming becoming obsolete, which lengthens the period over which which the cost of  9

product development can be amortized.

• Second, it speeds the development of product adaptations, which gives you the option to launch more products or to develop products that are more tailored to customer needs.

150

 

Product Development

• Third, it allows you to delay product development development efforts until you learn more about markets, competitor actions, and complementary technologies, which reduces your need to rely on long-term forecasts, and increases the likelihood that market conditions and customer needs will the same 10 when the product is introduced as when product development wasbe initiated. • Fourth, it gives your company the potential to develop the advantages of being the first to market.11 However, reducing product cycle time will not help your company if you do it  by cutting out important steps in the product development process, such as initial product design or market research (that will only cause you to produce undesirablee prod sirabl products ucts quickly). quickly).12 Rather, you need to reduce product cycle time while developing products that have the same features as those introduced more slowly. 13 Research has shown that several policies reduce product development time without reducing the quality of product attributes, including strong senior management support for rapid product development, an incentive structure designed to reward fast product development, focused project teams, powerful project leaders, experienced product development teams, autonomous product development efforts, 14

and strong external linkages. Therefore, you should adopting policies if reduced product cycle time is important to your consider competitive strategythese strategy. .  Key Points

• Companies want to reduce product cycle time to generate revenue sooner, respond faster to customer needs, have more time to gather information, and have a greater chance of creating a first mover advantage. • Product development time can be reduced by having strong senior management support for rapid product development, an incentive structure designed to reward fast product development, focused project teams, powerful project leaders, experienced product development teams, autonomous product development efforts, and strong external linkages. 15

C ONCURRENT  D EVELOPMENT As Figure 2 illustrates, you can reduce reduce cycle time by using concurrent development,16 which is the process of undertaking some product development steps simultaneously, rather than sequentially.17 For example, when you make the body of a car, a giant stamping machine called a die presses the metal into a shape of a part of the car, such as the door. In sequential design, automobile companies design the die first, then obtain the steel needed to make the parts, and finally begin cutting it. In concurrent development, the automakers get the necessary blocks of steel and start cutting them at the same time that the die is first designed.18 Concurrent development helps you to reduce cycle time because it does not require you to complete each step in the development process before you start subsequent steps.19 It also provides you with an early warning about potential problems at a later stage in development, while corrections can still be made upstream. 20 For

example, you might find that you cannot manufacture products at a desired quality level with a particular design. Because manufacturing engineers are working on manufacturing the product at the same time that product designers are finishing

151

 

Product Development

FIGURE 2

Concurrent and Sequential Development Sequential Development

Opportunity Assessment

Idea Generation Idea Generation

Opportunity Assessment

Prototype Development Prototype Development

Process Development

Process Development

Scale Up

Scale Up

Full Scale Operations

Full Scale Operations

Concurrent Development

Months of Development

This figure compares concurrent and sequential product development, and shows that concurrent development reduces product cycle time.

their designs, with concurrent engineering you can address manufacturability problems before the product design process has been completed.21 Concurrent development is difficult to implement because it imposes additional requirements on organizations. In particular, the overlap in the activities of manufacturing and design personnel demands a high level of communication between them. 22 Moreover, concurrent development requires the exchange of information between these groups to take place in real time, or its benefits will be lost.23 Consequently, to undertake concurrent development, you need to establish communication mechanisms that permit integrated problem solving across parts of your organization. Concurrent development is also risky. By undertaking steps of the development process before completing the prior stages of the process, you risk making product development decisions without adequate information. If your lack of information leads to poor decisions that require you to redo part of the development process, 24 you might have to spend much of your budget on something that would have been avoided with sequential development.25  Key Points

• Concurrent development, development, or the process of undertaking some product development steps simultaneously s imultaneously,, helps companies to reduce product cycle time. • However However,, it is risky and requires high levels of communication between parts of  the organization.

M ODULARITY

AND  P RODUCT   P LATFORMS

Modularity is the degree to which a complex product can be built from smaller

components (units from which something is made) that can be created independently, but function together.26 For example, cell phones are modular because they are composed of a number of components, such as the microchip and the screen.

152

 

Product Development

When products are modular, the functionality between the components is established through the creation of design rules and parameters that cover the product architecture, architectur e, interfaces between the modules, and standards for testing conformity to 27

design Therules. use of modularity will help you to accelerate your company’s pace of product development because each part of the organization responsible for the development of a module can conduct its design experiments concurrently with those of  other units.28 As a result, you can test your designs much faster than you can if you don’t take a modular approach. Modularity also offers several other benefits. It makes it possible to undertake very complex product development efforts. Take the case of Microsoft Windows. Over the years, Windows software has become too complex for designers to make major changes to it and still s till ensure that the software will work. Therefore, Microsoft Microsoft has created different versions of Windows by linking together thousands of tiny pieces of software code written by different engineers. As Windows has grown, this approach has caused the software to act in unpredictable ways, making it harder and harder to add new features. Moreover, Moreover, it made conducting daily tests of the interface of different pieces of the code impossible, requiring manual searches of thousands of  lines of code to find bugs in the system. To overcome this problem, Microsoft shifted to a modular approach to software development for its Windows Vista release. With a modular approach, different development teams were responsible for producing different parts of the software program, like graphics and spreadsheets, according to common specifications. The modules were then linked together according to a planned design. Modularity also permits you to customize your products to customer needs by allowing you to design in many alternative combinations of components. 29 This  breadth of alternatives, alter natives, in turn, allows you to target new market ma rket segments and to 30 respond more quickly to changing customer demands. Moreover, modularity makes product development easier to do. Because the smaller subsystems that compose a module are easier to develop than a whole nonmodular product, modularity reduces the expertise necessary for successful product development. Furthermore, modularity helps you to lower your cost of product development. Because you can use components that have worked in the past, your design costs are lower when you take a modular approach. And you can develop the components of  your product without costly coordination between the different teams developing them because changes in one component will not affect the functionality of another another..31 Perhaps most importantly, importantly, modularity allows you to purchase off-the-shelf components, which are often cheaper than custom-made ones. For example, take the cost savings that Teradyne, a scientific equipment manufacturer, achieved by adopting a modular approach to the development of a new semiconductor testing device. The device needed to have a spreadsheet feature for engineers to use when conducting tests on semiconductors. By using a modular approach to the development of the device, the company was able to purchase a Windows-based spreadsheet component for a few hundred dollars, rather than develop, on its own, a customized Unix-based spreadsheet at many times the cost. The latter approach, however, however, would have been necessary if the company had taken a nonmodular approach to product development.32

However, modularity has several drawbacks. First, as the personal computer  business illustrates, you often have to compete on cost when products are modular  because specialization by different component manufacturers drives down the cost

153

 

Product Development

of the overall finished product. Thus, when products are modular, the winner in competitive battles is often companies, like Dell Computer, that are the most efficient, not the most innovative. Second, modularity does notcomponents work well unless youcorrectly can define technical standards that ensure that different interface correctly. .33 In industries in which components are produced by different companies, this means that companies have to converge on a common set of technical standards for modularity to work. 34 And even when technical standards exist, the linkages between the components of  modular products sold by different companies are often worse than the linkages  between the components of modular products sold by a single s ingle company because their design and production involves less information exchange between the developers of the different components. For instance, Apple’s iPod device, iTunes software, and iTunes iTunes music store interface much more smoothly than the alternative that Microsoft and its collaborators offer because Microsoft’s products are composed of  modules made by several different companies.35 Third, the architecture of modular products is very inflexible because changing the architecture of a modular product requires convergence of the firms in an industry on a new technical standard. Therefore, modularity does not make sense when product architectures architectures need to be fluid and change from year to year year..36 So, should you adopt a modular approach to the design of your product? The answer depends. Research has shown that modularity makes the most sense when a product is composed of components that are very different from each other; when

FIGURE 3

A Computer Is a Modular Product

A computer is a modular product because it is composed of a set of smaller components—the components—the CPU, mouse, monitor, and keyboard—that are combined to produce a larger system.

154

 

Product Development

customer demand is heterogeneous; when technical standards have been established; when the pace of technological change is high; when the firms producing different components have very different capabilities; capabilities; and when the components do not 37

have to be highly integrated.

Product Platforms Many technology companies use a platform, or a common technological base to which different features are added, to create a family of products, each targeted at different customers. For instance, Intel has produced several microprocessor platforms, including the 8088, the 486, and the Pentium, off of which it has created many different d ifferent products.38 (T (Table able 1 shows Intel’s product product platforms and derivative products products during the 1990s and indicates how the derivative products varied on processor speed.) Product platforms can be built on a variety of different types of common technologies, including physical architecture, as is the case with automobiles; process technology,, as is the case with microprocessors; or a common set of components, as is technology the case with cameras.39 They can be created by changing existing products to con40

form to common as Lutron did with lighting control products, products.as However, However , they arecomponents, typically designed in advance to its support derivative Dell does with its computers. Using a product platform can benefit your company in a wide variety of ways. First, it allows you to offer greater variety to customers without straining organizational efficiencies. If you create different derivative products off of a common platform, you won’t have to conduct as much new product development or process redesign as you will if you create multiple products without the benefit of a platform. 41 You can also better coordinate resource use between products, plan for successive generations of products, and manage the timing of market entry. 42 Using a product platform may even lead you to invest more in designing your operations, which could lead to superior product architecture and better linkage between components.43 Second, using a product platform will reduce your costs. As long as the incremental costs of making derivative products are low relative to the cost of creating the platform in the first place, companies can benefit by developing a platform strategy. 44 Moreover, you can use a platform strategy to reduce manufacturing, materials, and procurement costs because products built off the same platform typically use common machinery,, parts, and other inputs,45 and change equipment setup less frequently machinery frequently..46 As a result, if you use a product platform, you will incur lower costs from changes to the setup and achieve better scale economies in purchasing inputs and raw materials. 47

TABLE 1

Intel Product Platforms This table shows the products that make up three different Intel product platforms: platforms: the 486, the Pentium, and the Celeron. 486

PENTIUM

CELERON

SX 16–25 MHz

Pentium 60–66 MHz

266 MHz

SL 20–33 MHz

Pentium II 233–300 MHz

300 MHz

DX2 20–66 MHz

Pentium III 350–600 MHz

533 MHz

Source: Adapted from Mohr, J., S. Sengupta, and S. Slater. 2005. Marketi ng of High Technolo echnology gy Products Products and Innovat Innovations ions (2nd edition). 2005. Marketing Upper Saddle River, NJ: Prentice Hall.

155

 

Product Development

Third, using a platform strategy helps you to segment the market on the basis of  price and performance because it helps you to develop different product variants for different groups of customers. This allows you to capture smaller market niches, as well as to hold on to customers as they develop interest in different different product features. For instance, many companies develop a low-cost product variant with the goal of  attracting low-end users. Then they offer product variants that provide greater performance on a variety of product dimensions, which gives them the opportunity to offer new products to customers as they become interested in the performance benefits that come from those product dimensions.48 As a result, a platform strategy allows companies to keep competitors from taking their customers by filling in gaps  between their product offerings. offerings.49 Fourth, using product platforms shortens product cycle time. If you don’t use a product platform, you will need to conduct additional R&D before you can introduce a new variant of your product. However, if you develop a product platform, you won’t have to conduct unique R&D for each new product that you introduce, but, instead, will be able to use the same development knowledge across across all of the products derived from the platform. As a result, you’ll be able to get your products to market more quickly if you use a platform strategy.50 However,, using a platform strategy has several costs. First, products created off  However of a product platform are less distinctive than products created independently. 51 Consequently, your company will not be well positioned to serve customers that have a strong preferenc preferencee for distinctiveness if it uses a platform strategy. Second, to use a platform strategy, you will need better cross-functional coordination in your organization. Because the designs of the different derivative products built off of a common platform need to be coordinated, developing products off of a platform is more complex and costly than the development of stand-alone products. 52 Third, with a platform strategy, strategy, lower-level products have to be “over designed” so that they can share systems and components with higher-level products.53 This “over design” makes low-end products from a platform more costly than low-end independent products, products, and this greater cost results in either higher prices and lower sales, or reduced profit margins, on those items. Given the pros and consare of more product platforms, should you use them? In general, product platforms useful if your when company operates in an industry that is not very dynamic but is very capital intensive, such as aerospace or automobile manufacture.54 In addition, platforms are more useful when you can achieve scale economies at the platform level, when the likelihood of overdesigning lower-end products is low, and when your markets are neither extremely homogenous nor very heterogeneous. When markets are homogenous, creating one standard product is often better than using a product platform, while focusing on the most attractive niche is often the best approach when markets are very heterogeneous. 55 The disadvantages of product platforms also lead many companies to combine modularity with product platforms to balance efficiency with flexibility when they develop new products. products. A standardized product platform platform with modular components provides the efficiency advantages of standardization and scale economies at the platform level, with the variation at the product level necessary to meet customer needs. For example, combining a standard platform with a modular product allowed

Sony to create 75 different models of the Walkman just by offering different combinations of standardized common modularized components, such as color, size, music format, and so on.56

156

 

Product Development

 Key Points

• Modularity speeds product cycle time, facilitates customization, makes product development easier, and reduces its cost. • However However,, modularity creates cost-based competition, makes product architecture inflexible, and requires technical standards to be developed. • Modularity makes the most sense when a product is composed of components components that are very different from each other, when customer demand is heterogeneous, when technical standards have been established, when the pace of technological change is high, when the firms producing different components have very different capabilities, and when the components do not have to be highly integrated. • Companies use platforms, or a common technological base for a family of products, to offer more variety to customers without straining organizational efficiencies, to facilitate market segmentation, and to reduce product cycle time; however,, the use of platforms reduces product distinctiveness, increases however demands for organizational coordination, and leads to the “over design” of  low-end products.

I DENTIFYING

THE  R IGHT  P RODUCT   F EATURES

You need to develop products with features—or characteristics of the product —that are better at meeting customer needs than the features of competitors’ products, or your competitors’ products will be more appealing to potential customers than your products.57 Take, for example, the problem faced by Philips with its home entertainment system. This product failed to sell because it was complicated, costly, costly, and difficult to operate in a market where competing products were simple, simple, inexpensive, and easy to operate.58 Unfortunately, you can’t ensure that your product will appeal to customers by including every possible feature that customers might want. If customers don’t want those features, their inclusion will impose unnecessary costs, which will require you to raise the price of your product and make it less appealing to customers. Moreover, Moreover, as the opening vignette to the chapter indicates, adding unnecessary features can complicate your product, also making it less appealing. As Figure 4 indicates somewhat humorously humorously,, to be successful, you need to identify and include just the features that will make your product attractive to customers. This is not easy because customers will often express preferences preferences for hundreds of different features if you ask them what they want. Because these preferences will not be of equal importance and will not be expressed by equal numbers of people, you will need to assess their importance and prioritize them.59 The next section describes several ways to identify and prioritize customer needs and to select product features that satisfy those needs.

Conjoint Analysis Conjoint analysis is a statistical tool that allows you to assess the relative importance to customers of differe different nt product features (including price) and to identify the

 best combination combination of features to meet their needs, even when when customers are not not aware

of the value that they ascribe to those features.60 Conjoint analysis is particularly useful when you need to make trade-offs between different different product attributes, and you want to see the relative contribution of those attributes to customers’ preferences.61

157

 

Product Development

FIGURE 4

Getting the Features Right

Not all companies develop new products with the right set of features to meet the needs of customers. Source: The Cartoon Bank. © The New Yorker Collection, 1998, Michael Maslin from Cartoonbank.com. All Rights Reserved.

You can even see the relative contribution when different factors interact or when preferences prefere nces are not linear, as is the case with the relationship between two brands. For example, suppose you want to know how important each of the following features of a laptop computer is to potential customers: speed, warranty length, 62

price, screen size,they and don’t weight. Potential customersfeatures might not able to tellBut youwith this directly because think about individual of be a computer. conjoint analysis, you don’t need customers to know how they feel about each of  these features. Customer preferences can be inferred from choices that they make  between products products with different different combinations combinations of these features. features. In a typical conjoint analysis, a factorial design is used. A factorial design presents all different combinations of product features to potential customers. For example, a new computer might have two dimensions of interest to customers: high and low price and laptop versus desktop. A two-by-two factorial design would would present all four computer combinations: high-price laptop, low-price laptop, high-price desktop, and low-price desktop. In a conjoint survey, potential customers are asked questions about their preferences for different versions of the product. (For example, they might be asked how likely they would be to purchase each version of the product on a scale of 0 to 100.) Once the data on these preferences are collected, regression analysis is used to identify the contribution of each product feature to the overall choices. That is, the regression weights provide a

quantitative measure of the importance of each feature to customers. 63 For example, suppose, you work for Michelin, and you want to know how much your brand name affects the likelihood that customers would buy your mountain

158

 

Product Development

 bike tires rather than the mountain mountain bike tires tires made by competitors. (The information in this paragraph is taken from the Sawtooth Software mountain bike simulation.) You know that mountain bikes have five important features: brand name, tread type, tread wear, weight, and tire type. You can collect data from respondents, asking them about their preference for products that represent different combinations of these five attributes. For example, you might ask, “on a scale of 0 to 100, how likely are you to buy a Tioga mountain bike with very firm, off-road treads that last 2,000 miles, and weigh 2 pounds?” Then you might ask, “On a scale of 0 to 100, how likely are you to  buy a Michelin mountain bike with very firm, off-road treads that last 2,000 miles, and weigh 2 pounds?” Then you might continue, “On a scale of 0 to 100, how likely are you to buy a Tioga mountain bike with very firm, off-road treads that last 3,000 miles, and weigh 2 pounds?” (Of course, in reality, you need more than just three questions. Because you have five attributes of the product, each of which has three possible categories, a full factorial design would have 125, or (5 3), combinations, requiring you to have a conjoint survey with 125 questions in it.) Figure 5 shows the output from a conjoint conjoint analysis on this product produced produced by a company named Sawtooth Software. Sawtooth has already conducted the conjoint survey and has inputted the data from it. This allows us to conduct the analysis and see what kind of results we get. We can use the “market simulator” feature in the software s oftware to create a scenario in which all of the products are the same on all dimensions, except that one is made by Michelin, another by Tioga, and a third by Electra. We can then have the computer run a regression analysis on the market research data by selecting “purchase likelihood.” The software then runs a conjoint regression analysis and produces the output shown in Figure 5 for the purchase likelihood likelihood and standard error based on the preferences given by respondents to the conjoint survey. (See the tire simulator at Sawtoo Sawtooth th Software Software for the actual actual analysis, which which can be downloaded downloaded at www.sawtoothsoftware.com/downloads.shtml). The analysis shows that customers prefer the Tioga brand the most, followed by Michelin and then Electra. However, the differences are not huge. The Tioga brand only makes customers about 2 percent more likely to buy a mountain bike tire than the Michelin brand. Moreover, you cannot even be sure, with 95 percent confidence, that there is any difference between the brands because of the size of the standard error. FIGURE 5

Conjoint Analysis Input and Output

Input Product 1 Product 2 Product 3

Brand Michelin Tioga Electra

Type 1 1 1

Tread 1 1 1

Weight 1 1 1

Wear 1 1 1

Output Product 1 Product 2 Product 3

Purchase Likeli lih hood 35.61 37.77 31.78

Standard Error 3.68 3.50 3.64

The purchase likelihood estimates in this conjoint analysis output show that customers are more

likely to purchase a mountain bike tire produced by Tioga than an identical tire produced by Michelin or Electra. Source: This output was generated from the Sawtooth Software tire simulator, which is available as a demo from http://www.sawtoo /www.sawtoothsoftw thsoftware.com are.com/downl /downloads.sh oads.shtml. tml.

159

 

Product Development

While conjoint analysis is a very useful tool, it is subject to several s everal important limitations. First, it can only be conducted if a product or service can be specified as a collection of independent attributes. (Otherwise, you can’t disentangle the relative contribution of different product features to potential customers’ preferences.) Therefore, conjoint analysis is more useful for functional products, like computers, where different product features are observable, than image products, like perfume, where they are not. Second, the respondents must be familiar enough with the product category to express their preferences. If your product is so new that potential customers do not understand it, conjoint analysis will yield poor results. This means that conjoint analysis is not a very effective technique for developing new products in new markets, and works much better for developing new products in existing markets. Third, you need to know what product attributes are salient to potential customers. Conjoint analysis only compares the relative importance of product attributes, and cannot be used to identify attributes to consider in the first place. Moreover, Moreover, you can only present so many combinations of attributes to potential customers  before fatigue sets in, so there is a limit to the number of attrib attributes utes that you can include in a conjoint analysis.64

Kano Method When you develop new products, you also need to consider the functional form of  the relationship between product features and the satisfaction of customer needs. Conjoint analysis assumes that this relationship is linear—the more of the feature the greater the satisfaction of customer needs—but needs—but it need not be. One feature might  be absolutely abs olutely necessary to satisfy customer needs, while another might be nice to to have but not essential. Another might delight customers if it were present but not affect their satisfaction if it were absent. Finally, a feature might be something to which customers are completely indifferent. indifferent. Take, for example, a hand sanitizer. It is absolutely necessary for the product to kill bacteria. If it fails to get rid of the E. coli, it is of no use to anyone. But dead is dead, so including a feature that makes bacteria “more dead than dead” will not satisfy customers any more than a feature that just kills bacteria. On the other hand, it isn’t absolutely necessary for a hand sanitizer to smell nice. People might prefer it if  it smelled good, but if it smelled like chemicals, it would still satisfy their needs. The mechanism that is used to get the chemicals into their container won’t affect customer satisfaction; customers are indifferent to that. In the end, cost might be the only feature that has a linear relationship with customer satisfaction. That is, the less the product cost, the more customers would prefer it. The Kano method is a technique for evaluating customer preferences when those preferences are different if a feature is present than if the feature is absent. 65 The Kano method is similar to a conjoint analysis in that it gathers data from potential customers about their preferences for different product attributes. However, it differs from conjoint analysis because data is gathered from potential customers about how both the presence and absence of a product feature affects their satisfaction. As a result, the Kano method can be used to differentiate product attributes 66

into As four categories: must haves haves, , delighters delighters, , linear satisfiers, indifferents. indifferents Figure 6 shows, must haves are those attributes thatand do not provide. any additional satisfaction to customers as performance becomes more functional, but whose absence makes customers dissatisfied. In contrast, delighters are those attributes

160

 

Product Development

FIGURE 6

Very satisfied

The Kano Metho Method d

Linear satisfier

Delighter

Fully functional performance

Must have

This figure is a visual representation of three of the four Kano categories: For “linear satisfiers” there is a linear relationship between customer satisfaction and functional performance; for “must haves” more performance does not increase satisfaction, but less performance reduces satisfaction; and for “delighters” less performance does not reduce satisfaction, but more performance increases it.

whose lack of functionality does not make customers dissatisfied, but whose presence will add satisfaction. Indifferents are those attributes whose functionality has no effect on satisfaction either positively or negatively. Linear satisfiers are those attributes whose lack of functionality decreases satisfaction and presence of functionality increases satisfaction. Using the Kano method will help you to identify “must have” and “delighter” features, which are often missed in conjoint analysis. Take, for example, the case of  the received message function on an e-mail program. A typical market research survey might ask “how would you feel if your e-mail program verified whether a message had been received?” Suppose that the average respondent answered this question with a score of 5 on a 5-point scale in which 1 equals “not interested” and 5 equals “very interested.” You might think that you need to put the feature in your product because respondents respondents like it so s o much. However, now suppose you administered a Kano survey that also asked the respondents “how would you feel if your e-mail program did not have a feature that verified whether a message had been received?” and the average response was a 3 on the 5-point scale. The combination of the 5 points on the first question and the 3 points on the second question would would indicate that the product was a “delighter.” “delighter.” Customers would still be interested in the product if it lacked the feature but would  be very interested interested if the feature feature were were present. Similarly, suppose your market research survey just asked customers “how would you feel if your e-mail program verified whether a message had been received?” and the average respondent indicated a score of 3. You might think that the feature doesn’t matter much to customers (since it doesn’t make them like the product) and decide not to include the feature. This could be a wrong interpretation. interpretation. If you also administered a Kano survey that asked the respondents “how would you

feel yourreceived?” e mail program didaverage not have a feature thata verified message had ifbeen and the response was 1 on thewhether 5-point ascale, you would find out that the feature was a “must have” and would need to be included in your product to satisfy your customers.

161

 

Product Development

Perceptual Mapping To know if your new product meets customer needs in a significantly better way than existing alternatives, you need to compare the features of your product with those of your competitors. This can be done by collecting information from potential customers about their perceptions of your competitors’ products as well as your own. This information can be presented in a perceptual map, or a visual display of  the perceptions of customers about the different features of competing products. On a perceptual map, competing products are displayed on the different dimensions that potential customers have identified as important to them. By looking at the lines for different products on the map, one can tell whether one product is perceived as better than another at meeting customers’ needs, and what features would need to be changed for your new product to be perceived as better than those of  your competitors. For example, Figure 7 shows a perceptual perceptual map in which customers clearly perperceive product 1 as better than its competitors on the dimensions that customers have identified as important to them: price, size, speed, cost, and functionality. This perceptual map indicates that products 2, 3, and 4 will not be able to attract customers away from product 1, given the current product attributes. Based on this information, companies developing products 2, 3, or 4 should redesign their products or decide not to enter the market.

Concept Testing Concept testing is a procedure in which customers are presented with a new product and are observed for, or asked for, their reaction. 67 Concept testing is valuable  because companies would like to know if customers customers are positively disposed to their product concepts before they go through the time and expense of introducing their products to the market. For example, a maker of outdoor grills might conduct a concept test to determine if a new grill concept that emphasizes design over function would be appealing to customers. Concept tests come in two varieties. A positioning concept test presents the product concept along with mock advertisements; whereas a core idea concept test

FIGURE 7

An Example of a Perceptual Map

7   s    t   c   u    d   o   r    P    f   o   s   g   n    i    t   a    R

6 5 Product 1

4

Product 2

3

Product 3

2

Product 4

1 Price

Size

Speed

Cost

Functionality

By presenting the results of market research visually, visually, in the form of a perceptual map, market researchers can tell which, of a set of competing products and services, are perceived as better by customers; in this case, Product 1 is clearly preferred by customers on all of the relevant dimensions.

162

 

Product Development

presents the idea without a marketing message. While the former approach offers a  better prediction prediction of customer reactions, reactions, it confounds confounds the concept with the marketing message. In contrast, the latter approach avoids the confounding of positioning, but it doesn’t predict customer reactions very well.68 Concept tests have a very important limitation for you to think about. If potential customers rate your product concept poorly, poorly, the concept test tells you not to develop the new product. But it doesn’t tell you anything about how to fix the product concept to make it acceptable to customers.69  Key Points

• To satisfy a customer need, you need to come up with a product that has features that are better at meeting that need than those t hose of competing products; this is not easy to do because simply including every feature that customers might  be interested in having will make your product product too expensive expensive and complicated complicated to be appealing. • Conjoint analysis is a statistical s tatistical tool that allows you to assess ass ess the relative importance of different product features to customers by looking at their preferences for products with different combinations of features. • The Kano method examines the relationship between between the presence of a product feature and customer satisfaction when that relationship may not be linear. • A perceptual map is a visual display of the perceptions perceptions of customers about the different features features of competing products; it facilitates the identification of the features that make one product better than another at satisfying customer needs. • Concept testing is a procedure in which customers are presented with a product idea and are asked for their reaction; it provides information about whether a company should move forward with the introduction of a new product.

P RODUCT   D EVELOPMENT T OOLS Many new products never reach the market,70 and more than half of all new products introduced to the market generate negative economic returns.71 This low success rate is the result of the complexity of developing a new product. To To have a successful new product, you must develop something that is technically feasible, meet the needs of customers in a much better way than existing alternatives, develop and produce the product in a profitable manner, manner, satisfy the demands of relevant stakeholders, and price the new product correctly. Regardless of whether you are running a start-up or an established company, you can increase your company’s success at new product development by using product development tools. (However, limitations on time and money mean that entrepreneurs entrepre neurs need to balance the value of using these tools against the financial and time constraints that they face.) Web-based tools help you to gather information about customer needs that you might otherwise miss. The house of quality helps you to ensure that your new product has the features that your market research indicated

that customers value. Prototyping and stage gates help you to ensure that your new products don’t fail late in the development process when the cost of failure is very high. Finally, design for manufacturing helps you to ensure that you can actually manufacture any new products that you design.

163

 

Product Development

Web-Based Tools Companies are beginning to use Web Web searches to figure out what products should be offered to customers and what attributes those products should have. For instance, National Instruments Corporation, which makes computer hardware and software for engineers, modified its products to have USB interfaces when it learned that engineers conducting Web Web searches for hardware tended to include “USB” in their search queries. The company also made a product for testing auto quality that was compatible with a particular data protocol based on the results of a Web search. Especially for entrepreneurs, who often have limited budgets for gathering information about desired product attributes, looking at Web search patterns provides an inexpensive way to access a large amount of data about customer preferences.72

House of Quality To ensure that your new products have features that meet the needs of your customers, you need to create specifications, or rules about how you will get desired product features. For example, a customer might have a need to “record her child’s  birth.” This need might lead a product developer developer to create a digital camera with the feature of 24 hours of battery life. This feature might lead to a specification of a Li-ion rechargeable battery.73 The house of quality is a product development tool that helps companies to specify product attributes that satisfy customer needs. 74 Specifically, it is a visual matrix that compares weighted product attributes with weighted customer needs, providing a basis for an improved conversation between marketing and design personnel about those things.75 Thus, the tool helps you to avoid developing new products with attributes that have no purpose in meeting customer needs, and to develop all of the product attributes that are necessary to satisfy customer needs. 76 For example, Figure 8 shows a house of quality for for a software product. product. The figure shows that customer needs for an “on time” and “high quality” software product correspond to certain software design and subcontractor management techniques. On the left side of the house of quality is the list of customer needs. Along the top of  the of quality arefitthe attributes the product. The intersections of theof rows andhouse columns show the between theof needs of customers and the attributes the product, which are weighted by the company’s assessment of how well they do on the fit between those two dimensions. The triangle on the left side indicates the intersection between different customer needs and the company’s evaluation of the importance of those intersections; while the triangle on the top marks the intersection between different product attributes and the company’s assessment of the importance of those intersections. The line titled “direction of improvement” marks the company’s assessment of whether the company’s product development efforts are getting better or worse at matching customer attributes and product demands. Although the house of quality works better for less complex projects than for more complex ones,77 it is valuable in a wide variety of product development settings for several reasons. First, it helps you to make trade-offs between product attributes in an informed manner.78 Second, the house of quality helps make your designs coherent by showing you the complementary changes that you need to make

to satisfy customers or to create a competitive product.79 Third, it allows the voice of  the customer to be carried through to manufacturing because the outputs from one house of quality can be used as inputs into another. For instance, the weight of a car

164

 

Product Development

FIGURE 8

A House of Quality

Voice of the Company Enterprise Product Development Capabilities Business Qualility Qua ty Pro Proces cesses ses Capture

File: Example QFD Project BIP Date: 2/5/2006 15:32 Weight Strong Symbol 9 Weak Symbol 1 Medium Symbol 3 Larger the Better 0 Smaller the Better 0 Nominal the Best 0 Strong Negative –3 Negative –1 Strong Positiv Positive e 9 Positive 3

  s    d   e   e   n   r   e   m   o   e    t   c   s   n   u   a   c    t   r    d   o   e   p   e   c   m   x    I   e   r   r   e   o   m    t   o   e    t   e   s   m   u    t    C   a    h    t   s    l   a   s   o   p   o   r   p   e   v    i    t   c   e    f    f

  n    i   w     o    t     e   c    i   r

  p    i    h   s   r   e    d   a   e    l    /    t   n   e   m   e   g   a   n   a   m     m   a   r   g   o   r   p   y    t    i    l   a   u

   t   s   o   c   o    t   n   g    i   s   e    d  ,   e    l    b   a    i   r   a   v    t   n   e    d   n   e   p   e    d   n    i   n   a   s   a    t   s   o

   t   n   e   m   p   o    l   e   v   e    d    t   c   u    d   o   r   p   n   a   e

  n   o    i    t   a   z    i   n   o   m   r   a    h   e   v    i    t   a    i    t    i   n    i   s   s   e   c   o   r

Projec Pro jectt Man Manage ageme ment nt

   k   r   o   w   m   a   e    t   e    t   a   r   o   p   r   o

   t   n   e   m   e   g   a   n   a   m   r   o    t   c   a   r    t   n   o   c    b   u   s   e   v    i    t   c   e    f    f

   t   n   e   m   p   o    l   e   v   e    d   e   r   a   w    t    f   o   s   e   v    i    t   c   e    f    f

   t   n   e   m   e   g   a   n   a   m    k   s    i   r   e   v    i    t   c   e    f    f

  m   e    t   s   y   s    t   n   e   m   e   g   a   n   a   m   e   u    l   a   v    d   e   n   r   a   e   e   v    i    t   c   e    f    f

Technology Development

  y   g   o    l   o   n    h   c   e    t   e   g   a   r   e   v   e

  n   o    i    t   a   r   g   e    t   n    i    l   a   c    i    t   r   e

  s   e    i   g   o    l   o   n    h   c   e    t    d   e   n   o    i    t    i   s   o   p     e   r

   E    P    Q    C    L    P    C    E    E    E    E    L    V    P

Direction of Improvement   y Produce innovative solutions    t    i    l that work   a   u    Q Quality of Product    d   e    d Company that is open, honest,   n understanding of customer needs   a   m   e Effective customer contact    D   r   e   m Want products on-time   o    t   s   u Target program cost performance    C

   7    6    1  .    0    7    6    1  .    0    7    6    1  .    0    7    6    1  .    0    7    6    1  .    0    7    6    1  .    0

This figure shows a house of quality—a tool to match product attributes with customer needs—for a piece of enterprise software.

Source: http://en.wik /en.wikipedia ipedia.org/wi .org/wiki/QFD. ki/QFD.

165

 

Product Development

door needed to satisfy customers about crash safety might be used as an input into parts deployment, where the weight of the door is used to determine the properties of hinge parts that are sourced.80

Prototyping Effective product product development involves minimizing the losses that come from failed efforts to develop products that work or meet customer needs. Because failure is an inevitable part of doing something new, you cannot prevent it from occurring. But you can change the timing of when it happens. As Figure 9 shows, costs increase as companies move through the stages of prodproduct development. Therefore, if you can increase the proportion of failures that occur early—say, during the design phase—you can reduce the costs of failure. 81 Moreover, you can increase the value of learning from failure by increasing the amount of learning that occurs at the earliest stages of the product development process. One of the main ways that you can encourage early rather than late failure is through prototyping. Prototypes are approximations of actual products, including drawings, physical objects, and computer simulations.82 Although prototyping adds costs to the product development development process, and cannot completely remove the risk of  83 design errors, it provides two important benefits. First, prototyping allows companies to consider design alternatives before the actual product is developed, which permits design errors to be detected at an earlier stage of development, when the cost of fixing them is lower. 84 Second, prototyping helps firms to gather accurate information from their customers about their product preferences preferences by providing a focal point 85 for a discussion of the product’s design. Today, many prototypes take the form of computer simulations. In the typical example, product developers use computer-aided design to create prototypes of  products with different attributes, which can be tested through the use of computer simulations (see Figure 10).86 Computer-based prototyping offers four advantages over the development of  physical prototypes: lower cost and faster development, 87  better information about trade-offs that comes from using the information processing capabilities of computers to test more precise alternatives,88 the creation of full-scale versions that do not have to  be chang changed ed when scal scaling ing up actu actual al produ products cts to realreal-worl world d level levels, s, and the reu reuse se of of modmodels from other products, which reduces mistakes, and accelerates the pace of the product development process.89 For instance, Dassault Aviation used computer-based prototyping prototypi ng to design its Falcon 7X business aircraft. As a result, it didn’t need to create a mock-up or test plane and cut its product development development time in half to seven months.90

FIGURE 9

Identify Opportunity

The Product Development Process

Evaluate Opportunity

Design Product

Test Pilot

Develop Manufacturing

Test Market

Scale up Production

Cost = $1000

Cost = $10,000

Cost = $100,000

Cost = $1,000,000

Cost = $5,000,000

Cost = $10,000,000

Cost = $30,000,000

Cost increases as you move to later stages of the product development process.

166

 

Product Development

FIGURE 10

Computer-Aided Design

This figure shows an example of a computer-aid computer-aided ed design for a solid model assembly created in NX, a commercial CAD/CAM software package. Source: http://en.wik /en.wikipedia. ipedia.org/wik org/wiki/Image i/Image:Cad_cr :Cad_crank.jpg. ank.jpg.

GETTING DOWN TO BUSINESS

Prototyping at IDEO91

IDEO is a consulting firm that helps clients in a variety of industries to design their products. Among the products that they have helped to design are the Palm V, Steelcase’s Leap Chair, and Apple’s first computer mouse.92 When designing products, IDEO’s staff creates

another, to refine the product design, discussing each one with the clients and end users. The company views each prototype as an “experiment” that results in a “failure” on some dimension. Through these failures the designers learn what features to put into the products. 93

prototypes out of cardboard, clay, LEGOs, and foam so that their clients, IDEO personnel, and end users can all have a common point of reference when discussing products. Typically, an IDEO designer will develop an initial prototype of key product components to use as a “straw man” for all participants to criticize so that he or she can come up with a better design. Then, IDEO designers may make several prototypes, one after

IDEO’s management believes in making rough prototypes quickly. They think that if the designers spend too much time making detailed prototypes, they will become enamored of the prototypes and will not change them in response to feedback. They also believe in never going to client meetings without a prototype to ensure that in all discussions the designers and clients focus on the same things.94

Stage Gates As Figure Figure 11 11 shows shows,, a stage gate is a decision tree that permits evaluation of the option to continue a project at particular milestones. 95 When using a stage gate

process, you make go or no-go decisions at different milestones, called stage gates. These decisions are informed by answers to key questions about the technology, the market, and expected financial returns, 96 such as whether the estimated internal rate of return on the project at that point in time is greater than the cost of 

167

 

Product Development

FIGURE 11

Idea Generation

The Stage Gate Process 1st

2nd

3rd

4th

5th

Gate

Gate

Gate

Gate

Gate

Project Scoping

Prototype Development

Market Evaluation

Product Launch

Production Scale-up

The stage gate process, developed by Robert Cooper, is a conceptual map for bringing a new product idea to market; when projects meet milestones, called stage gates, they are allowed to move to the next stage of development. Source: Adapted from http://www.prod-dev.com/st /www.prod-dev.com/stage-gate age-gate.shtml .shtml..

capital.97 The typical stages include idea screening, building a business case, design and development, product verification and validation, and product launch.98 Many companies use stage gates to manage the product development process. For example, Activision, Activision, a maker of video games, uses a stage gate process to decide whether or not to put game ideas into production. The first stage of the process is called “concept review” and is designed to figure out if the game’s concept makes technical sense and can be sold. If the concept review is positive, the idea passes to the second stage, called “assessment.” In the assessment stage, the evaluators determine whether the design for the game makes technical sense and can be sold. If the game design and positioning are viewed positively, positively, then the idea passes to the “prototype” stage. At this stage, the game’s look and feel are examined, and the promotion and launch are considered. If the idea passes through this milestone, it goes to the “first playable” stage. At this stage, evaluators determine if the game development process is adhering to budget limitations, deadlines, and quality. If the milestones at the first playable stage are met, the game is sent to the “alpha” stage, at which point evaluators determine whether the game plays as well as it was intended. 99 While the stage gate process is more effective for innovation innovation efforts that emerge 100

from research than ones that dostage not, until it is,the in previous general, astage very has useful tool. Because it restricts investment at the next been completed, a stage gate analysis helps you to avoid undertaking activities for which insufficient information about upstream stages has been gathered. This is particularly important  because each progressive stage in the innovation process takes more time t ime and costs more money than the previous stage. As a result, using stage gates will help you to ensure that your innovation efforts achieve key milestones before you spend large amounts of money on them. 101

Design for Manufacturing Design for manufacturing is a set of rules that structure the new product design process to make products easier to manufacture, thereby reducing costs and improving quality.102 For instance, one design for manufacturing rule is to reduce the number of  components in a product by combining different features and simplifying assembly.103

If you use fewer components, your customers can benefit by purchasing fewer parts, having higher manufacturing yields (a result of fewer part failures), and by spending less time testing components.104 Another design for manufacturing rule is to design products for easy fabrication, which reduces labor costs and delays in production.

168

 

Product Development

A third rule is to design products products to simplify tooling and minimize minimize setup, which speeds production, and makes it less expensive to change production lines.105 A fou fourth rth 106 rule is to design products so that the costs of the parts themselves are reduced. The use of design for manufacturing is helpful to companies seeking to develop new high-technology products by increasing the fit between product attributes and customer needs, as well as by shortening product development time.107 For example, when NCR used design for manufacturing to develop an electronic cash register, it reduced both the time to produce its product and the cost of that production by over 50 percent.108  Key Points

• Web search tools provide a cheap and effective way to gather some types of  information useful for product development. • To ensure that new products have features that meet the needs of customers, many companies use the house of quality quality,, a matrix that compares product attributes with customer needs. • Prototyping helps companies to reduce the cost of failure at new product development by making it occur earlier in the development process. • Stage gates are a decision-making tool that restricts investment in the more costly stages of the innovation process until key milestones have been met. • Companies use design for manufacturing, a set of rules that structure the new product design process to make products easier to manufacture.

DISCUSSION QUESTIONS 1. Should companies companies seek to reduce product cycle cycle time? Why or why not? 2. What are the pros and cons of concurrent product development? Do the pros outweigh the cons? Why

disadvantages of using each of them to figure out disadvantages what attributes your product or service should have to meet customer needs? 6. What are the pros and cons of using Web-based Web-based

or why not? 3. Explain modularity.. What are the advantages and modularity disadvantages of developing products in a modular way? When should you develop modular products? When should you not develop modular products? 4. Why is managing a technology platform platform different than managing a single innovation? When is platform management most important? Why? 5. Compare perceptual perceptual maps, conjoint analysis, and the Kano method. What are the advantages and

tools to support product outweigh the cons? Why development? or why not? Do the pros 7. Compare prototyping prototyping and the house of quality. quality. How does each of these tools help you to develop a product that customers want? What are the drawbacks of each of these tools? 8. What are stage gates? How do they help you to manage product development? 9. What are some important “design for manufacturmanufacturing” rules? Why are these rules useful?

KEY TERMS Components: The units from which something is

Concurrent Development: The process of undertaking

made. Concept Testing: A procedu procedure re in which customers are are presented with a product idea and are observed for, or asked for, their reaction.

some product development steps simultaneously, rather than sequentially. Conjoint Analysis: A statistical tool tool that allows allows you to assess the relative importance of different product

169

 

Product Development

features and to identify the best combination of features to meet the needs of customers. Core Idea Concept Test: A concept test that that involves involves

Perceptual Map: A visual display display of how people perceive the different features of competing products. Platform: A common technological technological base off of which a

the presentation of the idea without a marketing message. Design for Manufacturing: A set of design rules rules that structure the new product development process. Factorial Design: A researc research h design that presents presents all different combinations of product features to potential customers. House of Quality: A matrix that compares compares customer customer requirements and product attributes, which is designed to improve communication between marketing and product design personnel. Kano Method: A technique for evaluating evaluating customer preferences when those preferences are different if a feature is present than if it is absent. Modularity: The degree to which a complex product can be built from smaller subsystems that can be cre-

family products can be created to meet a common marketofapplication. Positioning Concept Test: A concept test that that involves the presentation of the product along with mock advertisements. Product Cycle Time: A measure of how long long it takes to develop new products. Product Development: The creation of new products. Prototypes: Any kind of approximation of actual products across attributes of the product. Specifications: The rules about how you would design a product with certain features. Stage Gate: A decision tree in which a project project can be evaluated for the option to continue at particular milestones.

ated independently but function together.

PUTTING IDEAS

INTO PRACTICE

1. Web-Based Tools Many businesses are now using Internet search tools to identify attributes for the products that they sell. The purpose of this exercise is to familiarize you with the use of these tools and allow you to evaluate them. Go to the following Web sites: http:/ http ://inven /inventory.overture tory.overture.com/ .com/ d/searchinventory/suggestion, http://adlab. microsoft.com/ForecastV2/Keywordtrends Web.aspx, http:/ http: //trends /trends.goog .google.co le.com, m, and

inner tube tire), tread type (off-road tread, on-road tread, and off/on road tread), tire weight (2, 3, and 4 pounds), and tread wear (tire lasts for 1,000, 2,000, and 3,000 miles). You should first look at the questionnaire that has been presented to customers so that you can get a sense of the information that has been collected. In the Sawtooth software package, this means that you need to select the “Tires1.smt” simulation, and

www.keyworddiscovery.com/search.html. Type www.keyworddiscovery.com/search.html. key words into each of the sites (for example, type in “hybrid vehicle” and “fuel cell” vehicle). Now look at the results of those searches. How do the results help you to conduct product development? What are some pitfalls with using Web searches to gather information for product development? 2. Conjoint Analysis To do this conjoint analysis, you need to use a demo of some software. Go to www.sawtoothsoftware.com/downloads.shtml#ssi web and download Sawtooth Sawtooth Software’s Software’s SMRT v4.7 demonstration demonstratio n software. Once you have the software installed, open the demo and go to “file” to select “Tires1.smt.” This demo contains data from a survey of 30 customers about their preferences for different mountain bike tires. Customers have been

then go to “compose,” and pull down on “run questionnaire.” Answer the questions as they come up to get a feeling for the market research information that has been gathered. Once you are done with that, figure out what combination of attributes will create a tire with the highest purchase likelihood. (Hint: You You need to go to market simulator in the Sawtooth software. You need to create different scenarios that hold all other features constant and vary one feature to figure out which value is highest. Then, you need to create a scenario with that combination of features and select “purchase likelihood” under “simulation method.”) 3. Design for Manufacturing This exercise examines design for manufacturing. Your company is considering developing a new cordless, rechargeable,

asked about brand (Michelin, Electra, and Tioga), tire type (very firm puncture proof solid foam core, somewhat firm puncture proof solid foam core, less firm puncture proof solid foam core, and standard

 battery-powered drill. Before the product  battery-powered product designers designers come up with a prototype of the product, you have  been tasked with with establishing establishing several guidelines guidelines for them to follow to make sure that the product that

170

 

Product Development

they design is easy to manufacture. manufacture. Please specify five design-for-manufacturing rules for the product. (Hint: Think about rules that reduce the number of 

1. 2. 3.

parts, the assembly the product, makeof  testingfacilitate of the product easier, of reduce the difficulty servicing the product, use common parts and materials, and take a modular approach.)

4. 5.

NOTES 1. Corts, K., and D. Freier. Freier. 2003. The rise and fall (?) of  Palm Computing in handheld operating systems.  Harvard Business Business School Case, Case, Number 9–703–519. 2. Markides, C., and P. P. Geroski. 2005. Fast Second: How Smart Companies Bypass Radical Innovation to Enter and Dominate New Markets. Markets. San Francisco: Jossey-Bass.

17. Schilling and Hill, Hill, Managing the new product product development process. 18. Poppendieck, M. 2005. Tactics Tactics and benefits of concurrent software development. Concurrency Concurrency,, 14(2): 1–19. Revolutionizing ng Product 19. Wheelwright and Clark, Clark, Revolutionizi

3. Teague, P. P. 2000. Father of an industry. Design News, News, March 6: 108. 4. Markides and Geroski, Geroski, Fast Second: How Smart Companies Bypass Radical Innovation to Enter and Dominate New Markets. Markets. 5. Teague, Father of an industry industry.. 6. Ramstad, E. 2005. Cellphone Cellphone squeeze play: Key parts on one chip. Wall Street Journal, Journal, November 17: B3, B5. 7. Griffin, A. A. 1997. The effect of project and process process characteristics character istics on product development cycle time.  Journal of Marketing Marketing Research, Research, 34: 24–35. 8. Schilling, M., and C. Hill. Hill. 1998. Managing Managing the new product development process: Strategic imperatives. Academy tives.  Academy of Management Executive Executive,, 12(3): 67–81. 9. Schilling and Hill, Hill, Managing the new product product

Development.. Development 20. Eisenhard Eisenhardt, t, K., and B. Tabrizi. Tabrizi. 1995. Accelerating Accelerating adaptive processes: Product innovation in the global computer industry. Administrativ industry. Administrativee Science Quarterly,, 40(1): 84–110. Quarterly 21. Schilling and Hill, Hill, Managing the new product product development process. 22. Terwiesch, C., C. Loch, and A. De Meyer. 2002. 2002. Exchanging preliminary information in concurrent engineering: Alternative coordination strategies.  Management Science, Science, 13(4): 402–419. 23. Wheelwright and Clark, Clark, Revolutionizi Revolutionizing ng Product Development.. Development 24. Allen, K. 2003. 2003. Bringing New Technology to Market. Market . Upper Saddle River, NJ: Prentice Hall. 25. Terwiesch, Loch, and De Meyer, Exchanging pre-

development process. 10. Wind Wind,, J., and V. V. Mahajan. 1997. Issues and opportunities in new product development: An introducMarketing tion to the special issue. Journal issue. Journal of Marketing Research,, 34(1): 1–12. Research 11. Wheelwright, S., and K. Clark. 1992. Revolutionizing Product Development. Development. New York: Free Press. 12. Wind and Mahajan, Mahajan, Issues and opportunities in new product development. 13. Cohen, M., J. Eliashberg, and T. T. Ho. 1996. New product development: development: The performance and time to market trade-off. Management trade-off. Management Science, Science, 42: 173–186. 14. Kessler Kessler,, E., and A. Chakrabarti. 1996. Innovation Innovation speed: A conceptual model of context, antecedents, antecedents, and outcomes. Academy outcomes. Academy of Management Review Review,, 21(4): 1143–1 1143–1191. 191.

liminary information in concurrent engineering. 26. Schilling, M. 2000. Toward Toward a general modular modular systems theory and its application to interfirm product  Academy of Management Review Review,, 25(2): modularity. Academy modularity. 312–334. 27. Baldwin, C., and K. Clark. Clark. 1997. Managing Managing in the age of modularity. Harvard modularity. Harvard Business Business Review, Review, September–October: 84–93. 28.. Ibi 28 Ibid. d. 29. Schilling, Toward Toward a general modular systems theory and its application to interfirm product modularity. 30. Sanchez, R., and J. Mahoney. Mahoney. 1996. Modularity, Modularity, flexibility, and knowledge management in product and organization design. Strategic Management Journal, Journal, 17 (Winter Special Issue): 63–76. 31. Fleming, L., and O. Sorenson. Sorenson. 2003. Navigating Navigating the

15. Ibi Ibid. d. 16. Clark, K., W. Chew, and T. Fujimoto. 1987. Product development in the world auto industry. Brookings Papers on Economic Activity, Activity, 3: 729–771.

technology landscape of innovatio innovation. n. Sloan  Management Review Review,, Winter: 15–23. 32. Bower, J. 2005. Teradyne: Teradyne: The Aurora project.  Harvard Business Business School Case, Case, Number 9-397-114.

171

 

Product Development

33. Sanchez and Mahoney, Modularity, flexibility, and knowledge management in product and organization design.

54. MacMill MacMillan, an, I., and R. McGrath. 2002. Crafting R&D Project portfolios. Research Technology Management, Management, 45(5): 48–59.

34. Fleming Sorenson , Navigating the technology technology landscapeand of Sorenson, innovation.

55. Krishnan and Gupta, Apprdevelopment. Appropriateness opriateness and impact of platform-based product

35. Mossberg, W. W. 2006. In our post-PC era, Apple’s device beats the PC way. Wall Street Journal, Journal, May 11: B1. 36. Ernst, D. 2005. Limits to modularity: modularity: Reflections Reflections on recent developments in chip design. Industry and Innovation,, 12(3): 303–335. Innovation 37. Schilling, Toward Toward a general modular systems theory and its application to interfirm product modularity. 38. Robertson, D., and K. Ulrich. Ulrich. 1998. Planning Planning for product platforms. Sloan Management Review, Review, Summer: 19–30. 39. Christiansen, C. 1998. We’ve We’ve got rhythm! Medtronic Corporation’s cardiac pacemaker business. Harvard business. Harvard Business School Teaching Note, Note, Number 5-698-056.

56. Schilling, N. N. 2005. Strategic Management of  Technological Innovation. Innovation. New York: McGraw-Hill. 57. Cooper Cooper,, R., and E. Kleinschmidt. 1987. New products: What separates winners from losers? Journal losers? Journal of  Product Innovation Management, Management, 4: 169–184. 58. Schilling and Hill, Hill, Managing the new product product development process. 59. Griffin, A., and J. Hauser. Hauser. 1993. The voice of the customer. Marketing tomer.  Marketing Science, Science, 12(1): 1–27. 60. Jedidi, K., and J. Zhang. 2002. Augmenting conjoint conjoint analysis to estimate consumer reservation reservation price.  Management Science, Science, 48(10): 1350–1368. 61. Dolan, R. 1999. Analyzing consumer preferences. preferences.  Harvard Business Business School Case, Case, Number 9-599-112.

40. Farrell, R.,improve and T. T. Simpson. 2003.in Product design to commonality customplatform products.  Journal of of Intelligent Intelligent Manufacturing Manufacturing,, 14(6): 541–556. 41. Meyer Meyer,, M., and J. Utterback. 1995. 1995. Product development cycle time and commercial success, IEEE Transactions in Engineering Management, Management, 42(4): 297–304 42. Wind and Mahajan, Mahajan, Issues and opportunities in new product development. 43. Krishnan, V., V., and S. Gupta. 2001. Appropriateness Appropriateness and impact of platform-ba platform-based sed product development. Management ment.  Management Science, Science, 47(1): 52–68. 44. Mohr Mohr,, J., S. Sengupta, and S. Slater. Slater. 2005. Marketing 2005. Marketing of High Technology Products and Innovations (2nd edition). Upper Saddle River, NJ: Prentice Hall. 45. Meyer Meyer,, M. 1997. Revitalize your product lines

62. Ibid. 63. Ibid Ibid... Ibid 64. Ibid Ibid.. 65. Stein, E., and M. Iansiti. 1995. 1995. Understanding user need needs. s. Harvard  Harvard Business Business School Note, Note, Number 9-695-051. 66. Bayus, B. Forthcoming. Forthcoming. Understanding customer customer needs. Blackwell Handbook of Technology and Innovation Management. Management. Oxford: Blackwell. 67. Dolan, Analyzing Analyzing consumer preferences. 68. Ibid Ibid.. 69. Ibid Ibid.. 70. Schilling and Hill, Hill, Managing the new product product development process. 71. Page, A. 1991. PDMA’s PDMA’s new product development practices survey: Performance and best practices.

through continuous platform Technology Management Management, , 40(2): renewal. 17–28. Research 46. Robertson and Ulrich, Planning Planning for product platforms. 47. Meyer, M., and A. DeTore. DeTore. 1999. Product development for services. Academy services. Academy of Management Executive Executive,, 13(3): 64–76. 48. Meyer Meyer,, Revitalize your product lines through continuous platform renewal. 49. Mohr Mohr,, Sengupta, and Slater, Marketing Slater, Marketing of High High Technology Products and Innovations. Innovations. 50. Krishnan and Gupta, Appropriateness Appropriateness and impact of platform-based product development. 51. Robertson and Ulrich, Planning Planning for product platforms. 52. Gottfredson, M., and M. Booker. Booker. 2005. Finding

PDMA Internatio nal Conference. Boston, 15th MA:Annual OctoberInternational 16. 72. Delaney Delaney,, K. 2007. The new benefits of web-search Journal, February 6: B3. queries. Wall Street Journal, 73. Bayus, Understanding Understanding customer needs. needs. 74. Hauser, J., and D. Clausing. 1988. House of quality. quality.  Harvard Business Business Review, Review, May–June: 1–13. 75. Schilling and Hill, Hill, Managing the new product product development process. 76. Griffin and Hauser, Hauser, The voice of the customer customer.. 77. Schilling and Hill, Hill, Managing the new product product development process. 78. Griffin and Hauser, Hauser, The voice of the customer customer.. 79. Hauser and Clausing, House of quality. quality. 80. Ibid Ibid.. 81. Buggie, F. F. 2002. Set the “fuzzy front end” in

your innovation fulcrum. Wall Street Journal, Journal, December 20: B2. 53. Krishnan and Gupta, Appropriateness Appropriateness and impact of platform-based product development.

concrete. Research Technology Management, 45(4): 11–14. 82. Ulrich, K., and S. Eppinger. Eppinger. 2004. Product Design and Development.. McGraw–Hill/Irwin. Development

172

 

Product Development

83. Thomke, S. 1998. Managing Managing experimentation experimentation in the  Management Science, 44(6): Science, design of new products. Management products. 743–762.

96. Ibi 96. Ibid. d. 97.. Ibi 97 Ibid. d. 98. Cooper Cooper,, R., and E. Kleinschmid Kleinschmidt. t. 1991. New product

84. Ibid. d. 85. Ibi Ibid. Ibi d. 86. Schi Schillin lling, g, Strategic Management of Technological Innovation.. Innovation 87. Schilling and Hill, Hill, Managing the new product product development process. 88. Millson, M., M., S. Raj, and D. Wilemon. Wilemon. 1992. 1992. A survey of major approaches for accelerating new product development. Journal development.  Journal of Product Product Innovation  Management,, 9: 53–69.  Management Management. New York: 89. Afuah, A. 2003. Innovation Management. Oxford University Press. 90. Poppendieck, Tactics Tactics and benefits of concurrent software development. 91. Adapted from Thomke, S., and A. Nimgade. 2000. 2000. IDEO product development. Harvard development. Harvard Business Business School

processes at leading industrial firms. Industrial  Marketing Mana Management, gement, 20(2): 137–148. 99. MacCor MacCormack, mack, A., and E. D’Angelo. 2005. Activision: Activision: The Kelly Slater’s pro surfer project. Harvard project. Harvard Business School Case, Case, Number 9-605-020. 100. Schilling and Hill, Managing Managing the new product development process. 101. Schilling, Strategic Management of Technological Innovation.. Innovation 102. Ulrich, K., D. Sartorius, S. Pearson, and M. Jakiela. 1993. Including the value of time in design-formanufacturing decision making. Management making. Management Science,, 39(4): 429–447. Science 103.. Ibid 103 Ibid.. 104. Ramstad, Cellphone squeeze play. play. 105. Schilling and Hill, Managing Managing the new product

Case, , /en.wikipe Number 9-699-143. 92. Case http:/ /en.wikipedia.org/wiki/IDEO dia.org/wiki/IDEO 93. Thomke, S. 2003. IDEO product product development. development.  Harvard Business Business School Teaching Teaching Note, Note, Number 5-602-060. 94. Thomke and Nimgade, IDEO product development. 95. Walryn, D., D. Taylor, Taylor, and G. Brickhill. 2002. How to manage risk better. Research Technology  Management,, 45(5): 37–42.  Management

process. 106. development Ulrich, Sartorius, Sartoriu s, Pearson, and Jakiela, Including the value of time in design-for-man design-for-manufacturing ufacturing decision making. 107. Schilling, Strategic Management of Technological Innovation.. Innovation 108. Clark, K., and S. Wheelwright. Wheelwright. 1993. Managing 1993. Managing New Product and Process Development. Development. New York: Free Press.

173

 

174

 

Patents

From Chapter 8 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 175

 

Patents Learning Objectives Patents: A Vignett Vignettee Introduction Why You You Need Intellectual Property Protection What Is Patentable? What Can Be Patented? Novel, Nonobvious, and Useful First to Invent Nondisclosure Getting Down to Business: Patenting a Snowman Accessory Kit Expansion of What Is Patentable Design and Plant Patents The Parts of a Patent

Defining the Claims Who Can Apply? Using a Patent Picket Fences and Brackets Patent Litigation Patent Trolls Should You Patent?  Advantages of Patenting Patenting Disadvantages of Patenting Effectiveness of Patents in Different Industries Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives

After reading this chapter, you should be able to: 1. Explain

why firms can easily and quickly imitate most of their competitors’ products and services, and why firms need to obtain intellectual property protection.

2. Explain

what a patent is, and identify the characteristics of an invention that are necessary to obtain one.

3. Identify

the different types of patents, and explain what they

protect. 4. Identify 5. Outline

the key parts of a patent, and explain what they do.

the trends over time in the expansion of what is patentable, and discuss the pros and cons of these trends.

176

 

Patents

6. Define

patent infringement and explain how patent owners can use the legal system to enforce their patent rights.

7. Discuss

the benefits and limitations of patenting, and

explain when patenting makes the most sense.

Patents: A Vignette1  In 2001, J.M. Smucker Co. instructed its attorneys to send a letter to a small caterer in Gaylord, Michigan, named Albie Foods Inc., demanding that Albie cease making and selling crustless peanut butter and jelly sandwiches (PB&Js). Smucker’s attorney explained that Albie’ Albie’s s PB&Js infringed on Smucker’s Smucker’ s patent for the sealed crustless sandwich patent, which the jelly maker uses to protect its Uncrustables sandwich. Much to the chagrin of Albie’s management, Smucker’s Smucker’s does, in fact, own the patent on the sealed crustless sandwich, U.S. patent number 6,004,596, which it bought from the inventors of this sandwich, Len Kretchman and David Geske. As the abstract of the patent indicates, Smucker’s Smucker’ s has a monopoly right to “a sealed crustless c rustless sandwich for providing a convenient sandwich without an outer crust which can be stored for long periods of time without a central filling from leaking outwardly.” outwardly.”2 Patents protect things that are stated in their claims. As will be explained more fully in this chapter, a claim identifies the parts of an invention that others may not imitate. Smucker’s attorney pointed out to Albie that the patent for the sealed, crustless sandwich makes 10 claims, the first of which is “a sealed crustless sandwich, comprising: a first bread layer having a first perimeter surface coplanar to a contact surface; at least one filling of an edible food juxtaposed to said contact surface; a second bread layer juxtaposed to said at least one filling opposite of said first bread layer,, wherein said second bread layer includes a second perimeter surface layer similar to said first perimeter surface; a crimped edge directly between said first perimeter surface and said second perimeter surface for sealing said at least one filling between said first bread layer and said second bread layer; wherein a crust portion of said first bread layer and said second bread layer has been removed.”3 This claim means that no one else has the right to make a sealed, crustless sandwich with at least one filling and a crimped edge. Moreover, Moreover, if anyone makes, imports, or sells such a sandwich, Smucker’s Smucker’ s has the right to sue them t o collect financial damages. Albie decided to fight Smucker’s, charging that it did not infringe on Smucker’s Smucker’ s patent because the patent was invalid. Because sealed, se aled, crustless sandwiches have been popular in the Midwest since the nineteenth century ce ntury,, Albie claimed that the technology described in the patent wasn’t novel. And, if an invention can’t be shown to be novel, a patent on it is invalid. In the end, this patent dispute was never decided in the courts. As is often the case, both sides agreed to settle. But the lessons from the example are, nonetheless, instructive. Patents protect a wide variety of products. They provide the right for people to sue to collect damages if others make, import, or sell anything that uses the part of the invention protected by the patent claims without the written permission of the inventor. Consequently, as a technology entrepreneur or manager, you need to understand the role

of patents in technology strategy.

177

 

Patents

I NTRODUCTION Intellectual property is of great importance to many companies. In fact, at some public companies, intellectual property now accounts for as much as 70 percent of the value of the business.4 Because intellectual property is an important source of value in companies in high-technology industries, managing intellectual property is an important part of o f technology strategy. As a technology entrepreneur or manager, you need to understand the different legal mechanisms that you can use to protect your intellectual property, property, and you need to develop ways to employ those mechanisms to your advantage. There are four ways companies protect their intellectual property by legal means: through patents, trade secrets, copyrights, and trademarks. This chapter focuses on patents. The first section of the chapter discusses why companies need intellectual property protection. The second section identifies what is patentable. The third and fourth sections identify the parts of a patent, and describe how to use a patent, respectively. The final section helps you to decide whether or not you should patent your inventions. By reading this chapter, you will learn how to use patents to protect your intellectual property, and how to avoid violating your competitors’ patents.

W HY  Y OU  N EE D  I NTELLECTUAL P ROPERTY  P ROTECTION Before you can understand how to use patents, trade secrets, copyrights, and trademarks as strategic tools, you first need to understand why you need intellectual property protection at all. The answer lies in the components of a successful technology strategy. Introducing an innovative new product or service that meets the needs of customers is a necessary, but not sufficient, condition for success. Success also depends on protecting your product or service, or the way it is produced and sold, against competitors. your competitors, rather than you, will capture imitation the profitsbythat flow from Otherwise, your innovation. i nnovation. Why is the ability of competitors to imitate your products and services so problematic to your profitability as an innovator? The answer is simple. You need to earn a profit on the sale of those innovative products and services to recoup the investment that you made to develop them. Initially Initially,, when you introduce a new product or service, you’ll have a monopoly; no one else yet offers a product or service to meet the same market need. Your monopoly position allows you to charge high enough prices to generate the profit margins that you need to recoup your investment. Unfortunately,, any success that you have will motivate your competitors to copy Unfortunately what you are doing. If your competitors can come up with a product or service that meets the same customer need, or if they can undermine your advantages in producing or selling your product by copying how you do those things, then they can capture some of the profits that you are earning. To make matters worse for you, the more successful you are at the introduction introduction of the new product or service—and the less you

want to be imitated—the more motivated your competitors will be to imitate what you are doing. Your success makes it more obvious that competitors should imitate; and in many cases, it provides them with the information that they need to imitate your new product or service, or your method of producing and selling it.

178

 

Patents

If imitators are not stopped, they’ll undermine all of your profit. To produce their copies of your initial product or service, imitators need to get access to the same resources as you are using—the employees, the capital, and the raw materials. As a result, they bid up the prices of these resources, causing your profit margins to fall. Moreover, the imitators take away some of your customers. Each customer that they woo away from you drives down your revenues, further hurting your profit margins. Clearly, you need to stop your competitors from imitating your new products or services if you are going to be successful. Unfortunately, doing this isn’t easy. Most new products are simple to copy, particularly for large, established firms. One study  by Richard Levin and his colleagues showed that approximately half of the time, the average unpatented new product can be duplicated by between six and ten competitors, at less than half the cost of the original development.5 Another study, this one  by Edwin Mansfield, showed that, on average, one-third of new new products can be be imi6 tated in six months or less (see Figure 1). Companies can figure out how to imitate your new products and services in a wide variety of ways. Many new products can be reverse engineered, with your competitor’s technical staff simply taking apart your new product and figuring out how it works. Once their engineers figure out how your product works, it is often very easy for them to come up with another way to do exactly the same thing. 7 Your competitors can easily hire your employees as a way to learn what they know. Labor markets are free in most countries, and people often leave their jobs to go work for competitors. So your competitors could figure out how to imitate your products and services by offering a higher salary to your employees to get them to  jump ship. Then they can use the knowledge that your employees have developed in the course of their careers to create products and services that imitate yours.

FIGURE 1

The Amount of  Time It Takes to Imitate New Products

35    d   e    t   a    t    i

30

   I   m 25   s    t    d   c  o   u   i    d  r   e   o   r   P 20    P  e   m   w   e   i    T    N  e 15    f   h   o   t   n   e   i 10   g   a    t   n   e   c 5   r   e    P

0 18 Months

Imitation doesn’t take very long; on o n average, more than two-thirds of the time it takes competitors less than 12 months to figure out how to imitate an innovator’s innovator ’s new product. Source: Based on information contained in Mansfield, E. 1985. How rapidly does industrial technology leak out? Journal out? Journal of Industrial Economics Economics,, 34(2): 217–223.

179

 

Patents

Sometimes simply working on similar new products allows your competitors to figure out how to copy your new products or services. Most competitors are working on new products and services that are similar to each other, and just knowing that your company has figured out a way to, say, make a product smaller or add features to it, is sometimes enough to allow your competitors to come up with an imitative product or service on their own.8 Your competitors can often look at public documents and publications to figure out how to copy your new product or service. 9 Because engineers and scientists have strong expertise in the areas in which they work, they can often extrapolate from partial information obtained in public sources and figure out how to imitate your new products or services, just on the basis of information that you have made public.  Key Points

• Intellectual property protection helps you to deter imitation by competitors, which will undermine your profits from innovation. • Most new products and services are easy to copy, and a large number of firms are able to duplicate those products and services at a low cost. • Competitors can imitate your products by reverse engineering them, by hiring your employees, by working on similar projects, and by reading your publications and patent disclosures.

W HAT  I S  PATENTABLE? A patent is a government-granted monopoly that precludes others from using an invention for 20 years (for utility patents) in return for the inventor’s disclosure about how the invention operates. Patents are based on a fundamental trade-off. In return for showing others how an invention works, and thereby advancing the level of technical knowledge in a country, inventors receive monopoly rights to their invention forhave a period of time.set of effects on technological innovation. On the one Patents a complex hand, they provide people with an incentive to innovate. In the absence of the monopoly right provided by patents, inventors often would be unable to capture the value coming from their inventions and, therefore, be unwilling to develop or exploit them. Moreover, the disclosure that patents require makes it possible for other parties to learn from inventions and make further advances, which would not  be possible if the inventors kept the inventions secret. On the other hand, patents can deter technological innovation by making it difficult for others to reap commercial value from undertaking further developments in an area, given the inventor’s property rights. For instance, some observers believe that patents on genes deter follow-on innovation because because they give the patent holder too much protection, thus deterring others from developing genetic tests based on the initial inventor’s discovery. Patenting is an old, and established, form of intellectual property protection. protection. The

first uses of patents go back centuries; and patents were important mechanisms to protect basic inventions of the industrial revolution, such as the steam engine. The patent system is so important that, in the United States, it is enshrined in Article I of  the Constitution, and has existed since the birth of the nation.

180

 

Patents

Despite being around a long time, patenting appears to be increasing in importance. Since 1983, the number of patents granted by the United States Patent and Trademark Office (USPTO) has increased by approximately 5.7 percent annually. 10 Currently, approximately 350,000 patent applications are made to the USPTO each year, and approximately 200,000 patents are awarded. And inventors now spend in excess of $5 billion per year to obtain U.S. patents to protect their inventions. 11

What Can Be Patented? As Figure 2 shows, many brilliant brilliant business concepts concepts cannot be be protected with with patents because only the mechanisms for exploiting ideas can be patented, not the ideas themselves.12 For example, you can’t patent the idea of a fast-food restaurant drive-through window. window. All you can do is patent p atent a mechanism for exploiting expl oiting the idea, such as the window itself. Because they are not embodied in physical form, most services are difficult to patent.13 So you can’t patent courteous service, even if it provides your business with a competitive edge. The best that you can do is to patent the process by which that service is created, as would be the case if you developed a robotic employee that could be programmed to be courteous all of the time, even if it were having a  bad day. You also cannot patent laws of nature or any substances that appear naturally, 14 such as chemical elements, because the government thinks of nature, not the person discovering them, as the inventor. The best that you can do to protect a discovery of  a natural substance is to patent the mechanisms for obtaining it, such as the process of leaching iron from rock. 15

FIGURE 2

Not Everything Is Patentable

Many great inventions are not patentable. /www.cartoonstoc nstock.com. k.com. Source: http://www.cartoo

181

 

Patents

You can get a utility patent, which is given for new or improved products and processes, for one of four things: a process (such as a chemical reaction), a machine (such as a laser), an article of manufacture (such as a diskette), or a composition of  matter (such as a genetically altered bacterium).16

FIGURE 3

A United States Patent

The figure shows the first page of a U.S. patent for a device to grasp tissues from a box. /www.freepatentson ntsonline.c line.com/59799 om/5979960.pdf?s 60.pdf?s_id=919c _id=919c52aeb0632d 52aeb0632d7cfd9d864 7cfd9d86498edd018 98edd018c. c. Source: http://www.freepate

182

 

Patents

Novel, Nonobvious, and Useful Patents are only granted for inventions that the patent office determines are novel, nonobvious, and Theand USPTO defines anialinvention as “novel” if it exi hassting not  been previousl previ ously y useful. invented inve nted is not a triv trivial improveme impro vement nt on an existin g invention—an improvement that comes from exchanging one material for another, changing the size or shape of a device, increasing its portability, or exchanging one element for an equivalent other.17 The patent office deems an invention to be “obvious” if it is a clear next step in i n technological development to a person who is an expert in the field (for example, an electrical engineer would be considered an expert on electrical circuits) or if the elements of  the invention all were present in existing patents. 18 For example, J.M. Smucker Co. was denied a patent on its method of applying filling to its Uncrustables sandwich product  because  becau se the patent patent examiner examiner assigned assigned to the the case believed that the the concept concept of applying applying peanut butter on one slice of bread and applying jelly on another would be “obvious” to anyone trained in the art of making a sandwich. 19 For the USPTO to view an invention as “useful,” it has to work, have a use, and  be functional.20 So you can’t patent something like a piece of music, which isn’t functional. However, being “useful” doesn’t mean that an invention has to have commercial value; in fact, most patented inventions generate no financial returns. Take, for instance, U.S. patent 5,023,850, for a dog watch that moves at seven times the rate of  a normal watch.21 While this device works, is functional, and has a purpose, it has no commercial value. Perhaps the number of dogs who can tell time and have disposable income is too small for that.

First to Invent As Figure 4 indicates somewhat humorously, the U.S. patent system differs from the patent systems in all other countries (except the Philippines) because the United States awards patents to the first party to invent something, not to the first inventor

FIGURE 4

The First-to-Invent Rule

In the United States, if two inventors come up with the same invention, the patent is awarded to the inventor who can prove that he or she invented it first. Foundation . Source: Estate of Charles Addams. © Tee and Charles Addams Foundation.

183

 

Patents

to file for a patent. The importance of the first-to-invent rule can be seen in the case of the rotational wheel interface in Apple Computer’s iPod. Microsoft was the first company to file for a patent on this technology, which is currently the subject of a patent dispute between Apple and Microsoft. For Apple Computer to prevail in this dispute, it has to file a declaration to the USPTO, which states states that it invented the device before before Microsoft, and the USPTO USPTO has to determine, as a result of an investigation of the declaration and Apple’s records, that Apple’s invention predated Microsoft’s patent application.22

Nondisclosure In the United States, patents are only awarded for inventions that have not been offered for sale and have not been publicly disclosed, either in an open forum or in print, more than one year earlier.23 While experimental testing of your invention is not considered public use, you can’t advertise the invention, issue a press release about it, present it in a seminar, sem inar, or even offer a description of it i t at a trade show.24 In fact, in some cases, even showing an invention to your friends might constitute a public disclosure. So, to keep your invention secret, you need to have anyone who looks at the invention  beforee you file for a patent  befor patent sign a nondisclo nondisclosure sure agreement. agreement.25

GETTING DOWN TO BUSINESS

Patenting a Snowman Accessory Kit

Do you know how to build a snowman? Are you so good at it that you think your approach can be patented? Believe it or not, someone has already done it. In 1993, Robert Kenyon of Scranton, Pennsylvania, received U.S. patent number 5,380,237 for a “Snowman Accessory Kit.” According to the abstract of this patent, the invention is of an accessory kit “having a head, an

the upper torso body and the lower torso body, to enhance the appearance of the snowman; and retaining means on each of said articles having a plurality of  appendages, each of said appendages being circumferentially spaced from every other appendage and each of said appendages further being longitudinally spaced from every other appendage whereby when an

upper torso body and a lower torso body all made of  packed snow, which consists of a set of decorative articles. A structur structuree is for inserting each of the decorative articles into the packed snow of the head, the upper torso body and the lower torso body, to enhance the appearance of the snowman. Components are for retaining each of the inserting structures within the packed snow of the the head, the upper torso torso body and the lower torso body of the snowman, snowman, so as to prevent the decorative articles from falling off of the snowman.”26 Given the claims of this patent, no one else has the right to make or sell “an accessory kit for a snowman having a head, an upper torso body and a lower torso  body,, all made of packed snow  body snow, which comprises: a set

article is inserted into said snowman and turned, each of said appendages follows the same insertion path in a corkscrew-like manner, thus retaining said article in said snowman.”27 This means that if you go outside and build a snowman by taking a carrot and putting it in the head of the snowman for its nose and inserting some sticks clockwise into the torso for its arms, then you have violated Robert Kenyon’s patent. He can sue you and collect damages. To make a snowman in this way legally, you need to get a license from Kenyon. But, before you think Keynon has a blockbuster, pioneering, snowman patent, you need to check out the prior art that he had to cite to get his patent. You see, Kenyon doesn’t have the rights to a kit to make the hat for the snowman, which was patented by Brian

of decorative articles having a circumference and a longitudinal axis; means for inserting each of said decorative articles into the packed snow of the head,

Schwuchow of of Hobart, Indiana or a snowman feature feature and accessory system, which was patented by Phillis Lorenzo of White Marsh, Maryland. 28

184

 

Patents

FIGURE 5

Growth in Number of Utility Patents Issued

180,000 160,000 140,000 120,000 100,000 80,000 60,000 40,000

20,000 0         9         4         9         1

        3        5         9         1

       7        5         9         1

        1         6         9         1

       5         6         9         1

        9         6         9         1

        3        7         9         1

       7        7         9         1

        1         8         9         1

       5         8         9         1

        9         8         9         1

        3         9         9         1

       7         9         9         1

        1         0         0         2

       5         0         0         2

Year

Number of Utility Patents Granted

Since the early 1980s there has been a dramatic increase in the number of utility patents issued in the United States every year. Trademark demark Office. Source: Created from data downloaded from the U.S. Patent and Tra

Expansion of What Is Patentable Over time, the U.S. government has steadily expanded the types of things that can be protected by a patent, leading to a rise in the number of patents issued (see Figure 5). Since 1980, patents can be obtained on genetically engineered organisms, like mice, genetically altered substances, like yeast, and human genetic sequences.29 In addition, while mathematical formulas are considered to be natural phenomena and cannot be patented, 30 those formulas that are applied to a structure or process, as occurs in software, have been patentable since 1981.31 As a result, patents now protect a wide range of computer software, from training tools to investment and insurance systems to e-commerce payment mechanisms.32 Furthermore, since 1998, business methods like Amazon.com’s “One-Click” system, which allows repeat purchasers to make Internet purchases without reentering information, can be protected by patents.33 The expansion of what is patentable has raised a number of important issues that

you need to consider as you formulate your company’s technology strategy: 1. The increase in the volume of patent applications has created backlogs in the patent office, making the process of getting a patent less efficient than it used to be.

185

 

Patents

2. The growth in the patent “thicket” has resulted in a lot of cumulative and overlapping patents, increasing the number of patent disputes, as well as the rate at which firms license their patents to each other (cross-licensing). 3. The expansion into business method patents, which tend to be broader and more obvious than other patents, has raised questions about the degree to which innovation is being hindered by property rights. 4. The expansion into genetically engineered organisms has raised questions about whether patents block follow-on research and are making it difficult to come up with new medical and pharmaceutical innovations.

Design and Plant Patents There are two types of patents other than utility patents, which we have been discussing: design patents and plant patents. Design patents are given for the appearance of products.34 For example, U.S. patent number D339456 protects an ornamental design for a shoe sole. Design patents differ from utility patents because they have only one claim and protect a piece of intellectual property for only 14 ye year ars. s.35 Plant patents are given only for engineered plants that are reproduced asexually.36 An example is Tropicana’s patent for the varieties of oranges used in its Pure Premium orange juice. Like utility patents, plant patents protect a piece of intellectual property for 20 years and have multiple claims.  Key Points

• A patent is a government-granted monopoly that precludes others from from using an invention for a specified period of time in return for the inventor’s disclosure about how the invention works. • Three types of patents can be obtained: utility patents, which are given for new or improved products or production processes; design patents, which protect the appearance of a product; and plant patents, which protect plants that are reproduced asexually asexually.. • Patents cannot be obtained on natural substances or ideas; only a working process, machine, manufacture, or composition of matter that is novel, nonobvious, and useful, has not been disclosed either in an open forum or in print more than one year earlier, and has not been offered for sale, can be patented. • In recent years, the variety of things that can be patented has increased and now includes genetically engineered organisms, computer software, and business methods.

T HE  PA R TS

O F A   PATENT

Patents have two key parts: the specification and the claims. The specification is a description of how the invention works, and may include accompanying illustrations. The specification is what you must trade off in return for the monopoly right

tions. The specification is what you must trade off in return for the monopoly right that you receive. Its purpose is to allow those skilled in the relevant technical area to reproduce your invention.37 The other major part of a patent is the set of claims, or the statements that identify a particular feature or combination of features that are protected by the patent.

186

 

Patents

The claims are what indicate whether another patent infringes on your patent (that is, violates your monopoly right). As an inventor, you want to obtain a patent with broad scope claims (as long as you can enforce them) because a patent prevents imitation of only those things specified in the claims. The broader the scope of the claims, the harder it is for other firms to make changes to your invention and work around your claims. For example, DryDock Systems Inc. has a patent for floating dry docks made of connected plastic cubes. But instead of having a patent that covers just the cubes or their method of  assembly,, Jet Dock Systems Inc. has a patent on the process of drive-on docking. As a assembly result, the company’s patent bars competitors that use a different method of assem bly of a floating dry dock from imitating its product. 38 While broad scope claims are valuable for the reasons just described, you also need to ensure that your patent will be held valid if challenged, and that you will  be able a ble to prove infri infringeme ngement. nt. The broader the scope of your y our claim claims, s, and the better it protects your innovation, the more likely other people will be to challenge it, in the hopes that they can get rid of your monopoly right and do what you aree doin ar doing. g. Unfortunately (for the patent holder), patents with very broad claims are often difficult to enforce. The less specific the claims, the less easily judges and juries can interpret the words in them.39 In addition, patents with very broad claims that cover entire methods of approaching problems, rather than just specific products, are more likely to be deemed invalid. For instance, a federal appeals court recently invalidated a University of Rochester patent on Cox-2 inhibitors, saying that the patent’s claims were too general for the patent to be valid, causing the University of Rochester to lose its lawsuit against Pfizer Inc. for infringement of the university’s patent by its drug, Celebrex. So how do you know how strong your patent’s claims are? You need to look at the patent to see if part of the claims could be changed or dropped and still yield the same level of protection. For example, if a patent claims the process for using a particular adhesive for attaching two pieces of metal, but you could easily attach the two pieces of metal with another adhesive, the patent has weak claims. All a competitor has to do to get around your adhesive patent is to substitute a different adhesive for the one that you have claimed. Sometimes companies cannot get one or two broad claims, and, therefore, try to protect their inventions with a large number number of claims. As Figure 6 shows, sometimes the number of claims that inventors seek on their patents is quite large indeed. Pioneering patents—basic patents in a technical area on which a wide range of  inventions build—are a special case of patents with strong claims. Control of these patents is important because they can be used to generate royalty payments from a large number of users. For example, the holders of pioneering patents in genetic engineering earned hundreds of millions of dollars in royalties from a variety of  companies using genetic engineering techniques to make drugs. Similarly, NEC Corp. has made a great deal of money licensing its pioneering patents on carbon nanotubes, which are being used for fuel cell batteries in notebook computers, transistors, wide-screen televisions, and sensors.40

Pioneering patents and patents with broad claims are especially important to you if you are starting a company. New firms often lack other forms of competitive advantage when they are first established. Strong patents allow you to create the value chain for your new business and establish additional competitive advantages  before your new product or service is imitated by other firms.

187

 

Patents

FIGURE 6

Number of Claims Cell Cycle Inhibition Sites

Adenosine Deominose Inhibition Purine Ring Synthesis Inhibition Nucleotide Interconversion Inhibition

Pyrimidine Synthesis Inhibition Purine Synthesis

Pyrimidine Synthesis Ribonucleotide Inhibition

Ribonucleotide XMP Synthesis Inhibition

Inhibition of Dihydrofolate Reduction TMP Block

Deoxyribonucleotides DNA

DNA Synthesis Inhibition

RNA

DNA Damage

DNA Adduct Formation Protein Synthesis

DNA Intercalation RNA Synthesis Inhibition

Protein Synthesis Inhibition Enzymes

TMP — Thymidine Monophosphate

Microtubules Microtubule Function Inhibition

The patent application for the electrical device and anti-scarring agent shown here was originally submitted to the USPTO with 13,305 claims; however, however, the patent was issued with only about 50 of those claims. Source: http:/ http://www.gelsin /www.gelsing.ca/bl g.ca/blog/?p=10 og/?p=106. 6.

The more pioneering your patent, and the broader its scope, the more competitor firms you can deter from imitating your new product or service. For instance, Friendster recently obtained a patent on the method of searching for people on the Web as a function of their social ties. Because this patent is so broad, Friendster can use it to stop Facebook and MySpace from engaging in similar Web-based social networking activities, or at least to obtain licensing fees from them. Consequently, the patent will help the start-up, whose performance has lagged as a result of competition from these two companies.41

Defining the Claims The claims that you are allowed to make are limited by what previous inventors, whose patents have already been granted, have claimed. The initial arbiter of what

claims you can get is the patent examiner. To help patent examiners determine what claims should be granted, as an inventor, you have a duty to provide citations to previous patents whose technical art you build upon in creating your inventions. These citations limit your property right to only those things not claimed in previously cited patents.

188

 

Patents

To disallow claims, patent examiners must provide legal reasons for their actions. If an examiner denies a claim, an inventor can file a response to the examiner, and have that response evaluated. If the decision remains negative after this revision, then an inventor can appeal to the Board of Patent Appeals for a conference with senior patent examiners. This appeal can result in either a reversal of the examiner or a continued rejection of the patent claim. If an inventor fails to obtain positive decisions at the level of the Board of Patent Appeals, he or she can ask the Board of  Examiners to consider the issue.42 Although patent examiners are very good at determining what claims should be granted, they sometimes grant patents that are too broad or that overlap, particularly in new technical fields where examiners lack expertise. For example, many observers have criticized the USPTO for initially allowing very broad claims on genetic engineering and Internet business method patents. 43 These broad claims may have increased the cost of developing new products by creating a complex morass of conflicting claims. It is important to note how human the patenting process is. Patents are not an objective exclusion device created by machines. Rather, they are the result of interpretations by human beings and a complex As inventors a result, some issued patents make sense and others do not.negotiating Moreover,process. sometimes who deserve patents are denied them.

Who Can Apply? Only inventors can apply for, and be awarded, patents. If you employ, or contract for, work done by others, and would like to own the patents on any resulting inventions, you need your employees or contractors to agree in writing to assign those patents to you. Otherwise, Otherwise, U.S. law assumes assumes that employees and independent contractors own the rights to the inventions that they make. To facilitate this assignment process, many companies specify it in their employment agreements.44 The limitation on who can be awarded a patent raises an important issue for start-up firms. Inventors who found companies based on their inventions own the patents awarded for those inventions unless they assign them to their thei r companies. To protect themselves, investors in start-up companies typically require inventorfounders to assign those patents to their companies. That way, if the ventures don’t do well, the investors have access to the businesses’ intellectual property assets and can sell them to recoup some of their losses. Of course, this means that inventors may lose control of their own inventions if their start-ups, which use these inventions, are unsuccessful.  Key Points

• Patents have two key parts: the specification, which describes how the invention works, and the claims, which identify the features protected by the patent.

Inventors want the broadest scope claims that can be enforced. • To get a patent, inventors craft claims that do not infringe on those of previous patents. • Only inventors can apply for patents, which they do by applying to the USPTO.

189

 

Patents

U SI N G

A   PATENT

As a technology strategist, you also need to understand how to use patents effectively.. This section discusses two important issues: the use of multiple patents to protively p rotect technology and the use of the legal system to enforce patents.

Picket Fences and Brackets Although you might try to obtain a broad patent with strong claims to protect your invention, such patents are not always possible. p ossible. As a result, you might want to apply for more than one patent to protect your new product or service. For example, if you run a biotechnology company that has invented a new drug, you might try to patent  both the molecule itself and the process for producing it. If you cannot get a single broad patent to protect your invention, then building a picket fence of patents around the core invention can be a way to protect your new product or service. For example, when Gillette developed the Sensor razor, it obtained 22 different patents to protect the product, including patents on the blades, the handle design, and the packaging container. 45 Because your competitors’ ability to imitate your invention without infringing on your patents depends on coming up with a way to do the same thing as your invention in a way not indicated in your patent claims, then obtaining a variety of patents can deter imitation by closing off  alternative paths. Because companies build picket fences of packet protection around their inventions, other companies engage in bracketing —the process of keeping an inventor from using his or her invention by patenting around it—to counteract their efforts. For example, if your competitor competito r has obtained a patent on a filament for a new, higher intensity light, you can keep that company from using its filament patent by patenting ancillary inventions, such as a new bulb, new housing, new connections, and a new shade, thereby bracketing the filament patent with your own patents on the rest of the light.46

Patent Litigation Because patents only provide you with the right to sue to collect damages from others who infringe on your patents, successful strategies for exploiting patents invariably involve legal action. Therefore, intellectual property litigation is, and has always  been, an integral part of technology strategy. But, before we can discuss how you can enforce your patents, you first have to know what patent infringement is. Infringement occurs when someone to whom you have not licensed your invention makes, uses, sells, or imports something covered by the claims of your patent. 47 Your invention does not have to be duplicated in its entirety for your patent to be infringed. If any part of what is claimed is duplicated, infringement has occurred. Moreover, infringement can occur inadvertently as well as deliberately and can occur if one party induces another to build something that is covered by your patent.48

So how do you know if your patent has been infringed? i nfringed? Infringement occurs if  another invention does substantially the same thing, in the same way, with the same result, as a patented invention. For example, a new bicycle wheel would infringe on an existing bicycle wheel patent even if the second bicycle wheel was

190

 

Patents

not exactly the same as the first but performed the same function in a similar manner to the patented wheel.49 But a patent on the engine of a fuel cell powered car would not infringe a patent on a car powered by an internal combustion engine. Even though the two technologies both make the car go, the two kinds of engines do so in a very different way, way, one by turning hydrogen into water and the other by  burning gasoline g asoline.. Still unsure about what constitutes infringement? Take a look at Figure 7. Using the example of a bucket, the figure shows you what questions to ask to figure out if  another invention infringes on your patent. If you suspect that your patent has been infringed, you should take legal action immediately. If you wait too long to enforce your patent rights, then the courts will presume that you know about the infringement and do d o not care to take 50 action. After filing a patent infringement lawsuit, you can ask the court for an injunction to stop the alleged infringer from engaging in the actions that violate your patent. An injunction alone may be enough for you to achieve your goal of stopping the infringement. infringement. For instance, when Amazon.com sued Barnes and Noble for violation of its patent onBarnes its One-Click purchase method, Amazon.com obtained an injunction against and Noble, which had the effect of shutting down Barnes and Noble’s Express Lane purchasing system, and achieved Amazon.com’s Amazon .com’s objecti objective. ve.51 If you initiate a patent infringement lawsuit, the defendants are likely to fight  back by seeking seeking to prove that your patent patent is invalid. (If (If your patent patent is invalid, then no one can infringe it.)52 Patents will be declared invalid if the invention is deemed obvious to people trained in the relevant technical art, or if the patent holder is shown to have publicly disclosed or sold the invention more than one year before filing the patent application.53 For instance, eBay Inc. recently won an infringement lawsuit

FIGURE 7

Does Doe s It Infringe? Infringe?

Alleged infringing device

Claim

Present in Infringing Device?

(Claim 1) A device for holding fluid, comprising: a bucket

Yes

a handle from which said bucket depends; and

Yes

a cushioning means attached to said handle

Yes

To determine if a product infringes a patent (in this case a bucket), you look to see if what is claimed in the patent is present in the device of the alleged infringer. Source: Adapted from http://www.smithhopen.com/ip_litigation.asp /www.smithhopen.com/ip_litigation.asp..

191

 

Patents

against it by MercExchange LLC by showing that MercExchange’s e-commerce  business process patents were obvious technical improvements to people trained in the art.54 The United States Supreme Court recently made it easier to challenge the validity of someone else’s patent. A licensee can now challenge validity while still paying royalties to the patent holder, which protects the licensee against the risk of a countersuit for patent infringement. Because patent infringement can result in large damages, the previous requirement that the licensee stop paying royalties to challenge patent validity kept many small, start-up companies from challenging the validity of  established companies’ patents. They were simply afraid of a countersuit that would cause them to go bankrupt if they lost. 55 If you win a patent litigation lawsuit, you can obtain monetary damages and/or a permanent injunction that prohibits the infringer from using the patented technology. The amount of damages that you can obtain depends largely on the intent of  the infringer. If the infringement was not deliberate, then the penalty is very lenient, usually a reasonable royalty that might be based on what the infringer would have paid you had they licensed the invention in the first place, or the amount that you 56

lost because of the infringement. Moreover, if the infringement is unintentional, the courts may require you to license the technology to the infringer in return for royalties. It is a very different story if the infringement is willful. Willful infringement occurs if the infringer deliberately copied your patented idea, tried to conceal its effort, or acted in bad faith. 57 Then the courts impose triple damages. 58 Therefore, the compensation that you can be awarded if you win a patent litigation lawsuit can be extremely large. For example, the University of California and Eolas Technologies Inc. were awarded $565 million from Microsoft for the infringement of a software patent.59 The largest patent infringement penalty ever awarded went to Polaroid. In that case, Kodak had to pay $990.5 million for infringement of Polaroid’s instant camera technology.60 Moreover Moreover,, Kodak had to close clo se down a $1.5 billion bil lion manufacturing operation, lay off 700 workers, and spend $500 million to buy back cameras it had sold using the infringing technology technology..61 While the damages that you receive from winning a patent infringement lawsuit can be large, so can the costs of enforcing a patent. For example, Jet Dock Systems Inc., a company that has invented a floating dry dock, has had to engage in six lawsuits to protect its patents against infringement since its founding in 1993. The cost of enforcing the company’s patents—$1.2 million since founding— is large for a company that only does approximately $15 million in sales annually.62 This makes patent litigation a major strategic issue if you run a new and small company. It also takes a lot of time to enforce a patent. For example, it took Ron Chasteen, the inventor of a patented snowmobile fuel injection system, 11 years to win his patent infringement lawsuit against Polaris Industries. 63 Because new and small companies must devote a large portion of their revenues and management time to enforce their patents, large, established companies often test their willingness to do so by infringing the start-ups’ patents and running the

risk that they might have to pay damages. Many entrepreneurs simply run out of  cash or energy before they can prevail in the several-year process of enforcing a patent and get nothing, or settle for much less than the triple damages that they would be due if they proved willful infringement.

192

 

Patents

Patent Trolls Patent trolls are start-up companies whose business model is to buy up patents and

seek royalties through licensing. Unlike large, established companies, like Lucent, IBM, and Texas Instruments, that make hundreds of millions of dollars a year on licensing, patent trolls do not produce products. The main strategy for patent trolls is litigation. Because patents give inventors the right to sue others for making or using an invention without permission, entrepreneurs can assemble a portfolio of patents and use the threat of litigation to collect royalties from potential infringers. For example, the former chief technology officer at Microsoft, Nathan Myhrvold, has created a company that has purchased several thousand software patents and uses the potential of litigation to motivate infringers to license them.64 Many patent trolls buy patents cheap from the creditors of bankrupt technology companies. With those patents in hand, they seek licenses from users of the patented technology, generally asking for a small royalty from companies generating high revenues from the technology. For instance, they might seek a 1 percent royalty from a company whose product is generating $5 billion annually. Even though the trolls’ odds of winning patent infringement lawsuits are small, the potential payoff is high enough to make significant profits off of their investments in the patents even if they win only occasionally occasionally.. Often companies decide that it is better to settle with patent trolls than fight them in court. Patent trolls have a powerful strategic tool—the injunction. If they can convince a court to stop a potential infringer from producing its products or services until after an infringement lawsuit is decided, then the alleged infringer will often settle and pay a royalty to the patent troll, rather than shutter its business for months, or even years. For instance, NTP won a $612.5 million settlement from from RIM, the maker of the BlackBerry, BlackBerry, because NTP got a federal judge to issue an injunction against RIM that would have caused them to shut down BlackBerry service while the case was decided.65  Key Points

• Effective patenting strategy often involves the creation of a picket fence of  patents around a core invention and bracketing competitors’ patents. • Patents provide the right to sue others if they infringe on your patent by making, using, selling, or importing something covered by the claims of your patent. • If a company wins a patent infringement lawsuit, it can obtain as much as triple damages and an injunction prohibiting the infringing activity. • Triple damages are awarded for willful wil lful infringement, while lost l ost profits or imputed royalties are common penalties when infringement is accidental. • Because patents that are invalid cannot be infringed, a common defense against infringement is to invalidate the patent by demonstrating that the inventor had disclosed the invention prior to filing for the patent, or that the invention is obvious to a person trained in the art. • Because small and new firms must spend a large portion of their revenues and senior management time to enforce their patent rights, large, established firms

sometimes willfully infringe their patents, believing that the small, new firms will not have the money or energy to fight back. • Patent trolls are companies whose purpose is to buy up patents and enforce their claims through litigation or threat of litigation.

193

 

Patents

S HOULD  Y OU  PATENT ? As you might suspect, there are advantages and disadvantages to patenting. Given these pros and cons, you need to decide whether or not to patent your inventions. To help you make informed decisions, this section discusses some of the important advantages and disadvantages of patenting.

Advantages of Patenting Companies patent for a variety of different reasons. Table Table 1 presents data from a survey conducted by faculty members at Carnegie Mellon University, which indicates the major reasons why U.S. manufacturing firms say that they patent. As the figure shows, the most common reason, given by 96 percent of respondents, is to prevent copying.

Barrier to Imitation  Under certain circumstances, patents can be an important barrier to imitation and a powerful mechanism to capture the returns to innovation. 66 For example, the mobile telephone company, Qualcomm, was founded to exploit a technology called Code Division Multiple Access (CDMA) that allows more efficient use of  the radio spectrum for cellular telephones. Qualcomm obtained a patent on this technology and used the patent to protect its products as well as to earn revenues  by licens l icensing ing the techno technology logy to other o ther firms firms.. Qualcom Qu alcomm m has ha s grown gro wn into i nto a $3 billi billion on company, successfully competing against large, established firms, such as Motorola—something it would have have been unable to do if CDMA had not been 67 patented.

Legal Protection  Patents also help companies to use the legal system to protect their intellectual property. It is much easier to use the legal system to enforce patents than to enforce trade secrets. Moreover, having a patent helps you to defend your firm against patent litigation by allowing you to counterclaim in an infringement lawsuit. 68

TABLE 1

Why Companies Say That They Patent The table shows the proportion of R&D managers surveyed about their company’s approach to patenting who responded that their company patented for each of six reasons. REASON

PERCE ERCENT NT OF COMPANIES GIVI IVING NG THE REASON

Prevent copying Block others

96 82

Prevent lawsuits

59

Use in negotiation

47

Enhance reputation

48

Licensing 28 Source: Adapted from Cohen, W., R. Nelson, and J. Walsh. 2000. Protecting their intellectual assets: Appropriability conditions and why U.S. manufacturing firms patent (or not), NBER Working Paper, No. W7552.

194

 

Patents

Because patent litigation often results in court-mandated licensing or licensing from the settlement of lawsuits, the use of patents to protect intellectual property also has the benefit of creating a new source of revenue for many companies. In some cases, this source of revenue can be quite large. For example, much of IBM’s $1.4 billion annual royalty flow is the result of patent litigation.69

Value Chain Leverage  Patents also give companies control over other firms in their value chain. By owning patents that are used by your customers or suppliers, you can influence their behavior and make them act more favorably toward you. For example, Nokia has obtained patents on cell phone speakers even though it does not produce phone components  because the patents give the company leverage over its supplier of speakers. 70 Markets for Knowledge  Having a patent facilitates the sale of technology to other firms. While patented technologies can be sold to others, technologies protected by secrecy cannot. Therefore, to license a technology to another company, you need to obtain a patent on it. Take the of aapodiatrist in Bedford, New shoe” Hampshire, whoexample invented “comfortable high-heel (U.S. named patent Howard number Dananberg, 5,782,015). Because Dr. Dananberg lacked the assets necessary to manufacture and sell shoes that were attractive to women, he sought to license his technology to companies that already produced shoes. To do this, he needed to turn over his shoe designs to shoemakers, which necessitated obtaining a patent. Otherwise, if he had tried to license his invention, he would have had no protection against an unscrupulous licensee who chose to steal his designs.71

Raising Funds  Patents help new companies raise money because they provide a verifiable source of  competitive advantage. Investors can see the mechanism through which the new venture will deter imitation, reducing their uncertainty about the value of the venture. Moreover, patents offer salable assets if a new venture fails. Because investors can “take the patents to the bank” at the end of the day, they see their investments in start-ups with strong patents as partially collateralized, increasing the amount of  money that they are willing to provide to them. 72

Disadvantages of Patenting While patents are valuable for many of the reasons just described, they also have several disadvantages.

Effectiveness at Deterring Imitation  Patents are not always effective at deterring imitation. Sometimes other firms can invent around your patents. ( Inventing around is the process of coming up with something that accomplishes the same goal as the patented invention without violating the claims of the patent.) By inventing around your patents, other companies can use your invention without having to pay you royalties and without having to incur

the high costs of developing the invention. If your competitors can invent around your patents, then those patents aren’t worth the cost to get. Moreover, patenting might actually be doing you more harm than good. By patenting, you have to disclose the specifications of your

195

 

Patents

invention, which might be the source of information that your competitors need to copy your product. Inventing around occurs when there are multiple ways of accomplishing the same goal. For example, an imitator can often invent around an electronic device patent by changing the design of the circuitry, allowing the imitative product to satisfy customers in the same way as the innovative one but without violating the claims on the patented invention. Thus only for inventions for which there is a single way to accomplish a particular goal are patents very effective at deterring imitation. For example, the patent on Symantec Corporation’s antivirus software, which finds computer viruses without searching every byte of data, is effective because software engineers do not believe that there is any way other than Symantec’s to find computer viruses without checking all of the data in a file. 73

Benefits of Nondisclosure  Patenting is disadvantageous when a company will gain more from nondisclosure than from a government-granted monopoly. monopoly. A patent gives you a 20-year monopoly on your invention, but secrecy might allow that monopoly to last longer. For example, the chemical formula for Coca-Cola has been maintained as a secret for over 100 years. (Because the beverage is composed of complex natural substances, it is not possible to reverse engineer and duplicate it. 74) As a result, no other companies have  been able to create beverages that taste exactly like Coke. If the formula for CocaCola had been patented, its chemical composition would have been disclosed in the specifications of the patent. Once the monopoly on it expired decades ago, competitors would have been able to produce soft drinks with exactly the same chemical composition (and taste!) as Coke. Pace of Change  Patents are not very helpful when the pace of technological change is very fast. When technological change is very rapid, the inventions that patents protect quickly  become irrelevant. Given the time it takes to obtain patents, and the cost cost of patenting, you probably75won’t be able to earn sufficient payback to justify the investment in such patents. For example, suppose your semiconductor will be obsolete in two years because of the pace of innovation in that industry. You might be better off just keeping it secret and using that proprietary knowledge to ensure that you, and not your competitors, will be able to develop the next generation of semiconductors. If  you patent your invention, the semiconductor will be obsolete by the time that the patent issues. But once it issues, the invention will be public knowledge, and that might help your competitors to develop the next generation of semiconductors. Moreover, to obtain a patent, you need to explain how the invention works in a way that enables a person trained in the art to make the invention. Sometimes, this information makes it possible for others to leapfrog your invention. If that is the case, you are better off keeping the invention a secret.

Difficulty Proving Infringement 

Patents are not very useful when proving that others have infringed your patent, or defending it in a lawsuit, are very costly or difficult. If you cannot amass the evidence that it takes to prove that infringement actually occurred, occurred, or the fixed costs of defending a patent are so high that it does not pay to protect it through the court system, then

196

 

Patents

TABLE 2

The Advantages and Disadvantages of Patenting Patents have many advantages and disadvantages disadvantages that need to be considered when deciding whether or not to patent an invention. PROS

CONS

Helps to raise money

Not always effective at deterring imitation

Often slows imitation

Disclosure not always worth the 20-year monopoly

Facilitates markets for knowledge

May cause competitor leapfrogging when the pace of technological change is rapid

Offfers leg Of egaal prote tecction of int nteellec ecttual prop opeert rty y

Some mettimes cannot be enf nfo orced

Creates value chain leverage

you will not get enough of a return on your investment in a patent to justify its cost. 76 For example, it may be very difficult to prove that someone infringed on a surgical method that you invented, such as a type of incision used in cataract surgery, because there are no sales of physical likemight drugsnot or devices, whichyour would show that 77 As a objects, your method was used. result, you want to waste money trying to patent the surgical method.

Effectiveness of Patents in Different Industries Patents are not equally effective in all industries. In general, they tend to be more effective in industries in which the core technology is biological or chemical, and less effective in industries in which the core technology is mechanical or electrical. 78 Why? The reason has to do with the difficulty of accomplishing the same goals through different technical means. For mechanical or electrical devices, you can make slight modification to the design and accomplish the same goal, but you cannot do this with things that are biological or chemical. For instance, a drug has a very precise molecular structure, and slight alterations will often transform the drug from something something while relatively device’s major changes to the structure ofbeneficial electronic to circuitry oftenharmful, do not alter an electronic effectiveness. Researchers have examined how effective patents are in different industries. Table 3 provides some data, adapted from the Yale Yale survey on innovation, on this question. The table shows that in industries like drugs and chemicals, patents are very effective at protecting new products, but that in industries like cosmetics or pulp and paper, they are not. 79 These differences in patent effectiveness explain why obtaining patents is crucial to generating high financial returns in industries in which patents are very effective, like pharmaceuticals.80 It also explains why start-ups in some industries, like  biotechnology,, often specialize in technology development and do not build assets  biotechnology across the different parts of the value chain, relying, instead, on licensing to capture value from innovation.

 Key Points

• Patents slow imitation, facilitate legal protection of intellectual property, make markets for knowledge possible, enhance value chain leverage, and help new firms to raise rai se money.

197

  

Patents

TABLE 3

Effectiveness of  Product Patents by Industry The table shows the scores of different industries on a questionnaire given to R&D managers about the effectiveness of  product patents in their industry.

INDUSTRY

POINT NT SCA SCALE LE — 7 EQUALS PATENT EFFECTIVENESS (7-POI EFFECTIVE VE”) “VERY EFFECTI

Drugs Organic chemicals Inorganic chemicals

6.5 6.1 5.2

Steel mill products

5.1

Plastic products

4.9

Medical devices

4.7

Motor vehicle parts Semiconductors

4.5 4.5

Pumps and pumping equipment

4.4

Cosmetics

4.1

Measuring devices

3.9

Aircraft and parts Communications equipment

3.8 3.6

Motors, generators, and controls

3.5

Computers

3.4

Pulp, paper, and paperboard

3.3

Source: Adapted from Levin, R., A. Klevorick, R. Nelson, and S. Winter. 1987 Appropriating the Returns from Industrial Research and Development. Brookings Papers on Economic Econo mic Activity, 3: 783–832.

• However, patents allow the disclosure of information that helps competitors to invent around a technology, sometimes protect an invention for less time than secrecy, permit technology leapfrogging when the pace of technological change is rapid, and are not worth their cost when infringement is difficult to prove. • Patent effectiveness varies substantially across industries because of differences in the nature of technology, and these industry differences affect several aspects of technology strategy strategy.. DISCUSSION QUESTIONS 1. What can and can’t be patented? What What keeps some inventions from being patented? How have the criteria for patentability changed over time? 2. Patents on genes differ from patents on other other things in two important ways. First, if you patent a gene, there is no way for rivals to make something better, better, which is not the case on other technologies (e.g., semiconductors). Second, researchers cannot identify a gene that causes a disease and develop a

3. What are the different parts of a patent? What What purpose do they serve? 4. What strategic actions can you take to increase increase the effectiveness of your patents at deterring imitation? Why are these actions effective? 5. What actions should you take if you believe that that your patent has been infringed? Should you engage in litigation to protect your patents? Why or why not?

genetic test for it without without getting DNA samples

6. What are the advantages advantages and disadvantages of 

from people who have the gene, while researchers can patent other technologies without obtaining anything from other people. For these two reasons, many people think that we shouldn’t be able to patent genes. Do you agree or disagree? Why?

patents? What do these advantages and disadvantages suggest about the use of patents as part of  your technology strategy? 7. Why do patents work work better in some industries industries than in others?

198

 

Patents

KEY TERMS Bracketing: The process of keeping an inventor from using his or her invention by patenting around it. Claim: The part of a patent that states s tates what the patent precludes others from imitating. Infringement: The action to make, use, sell, or import something covered by the claims of a patent, without permission of the patent owner owner.. Invent Around: To come up with a solution that does not violate the claims of a patent but accomplishes the same goal as the patented approach. Patent: An exclusive right given by the government to preclude others from duplicating an invention for a

PUTTING IDEAS

INTO PRACTICE

1. Evaluating Patents Go to www.uspto.gov. Do a patent search and look up the following patents: number 5,173,051—“Curriculum Planning and Publishing Method,” number 5,241,671— “Multimedia Search System,” number 6,080,436— “Bread Refreshing Method,” number 5,443,036—“Method of Exercising a Cat,” number 6,368,227—“Method of Swinging on a Swing,” num ber 4,597,046—“Securities Brokerage Cash Management System Obviating Float Costs by Anticipatory Liquidation of Short Term Assets,” and number 6,467,180—“Use of a Tape Measure to Determine the Appropriate Bra Size.” Should the USPTO have granted these patents? Why or why not? (In answering these questions, think about whether the inventions are obvious, valuable, and useful, and whether they infringe on prior patents.) 2. Basic Patent Searches This exercise is designed to help you evaluate the potential to patent a piece of  intellectual property. property. Please follow the steps below to figure out if you can obtain a patent. Step 1: Identify an invention. (It can be your own or one that someone else developed. If you can’t come up with one on your own, ask your instructor to suggest one for you to investigate.) In two paragraphs, explain why you think that the invention is

NOTES

specified period of time in return for disclosure of an invention. Patent Trolls: Start-up companies whose business model is to buy up patents and seek royalties through licensing. Picket Fence: The set of patents that are obtained to offer a wall of protection around a core invention. Pioneering Patents: The basic patents in a technical area on which a wide variety of patents build. Specification: The part of a patent that describes how an invention works.

patentable. Make sure that you explain why the invention is novel, nonobvious, and useful. Step 2: List several key words that identify identify the invention. Now go to the U.S. government’s patent Web site at www.uspto.gov www.uspto.gov and search for patents with the same key words as your invention. List all of the patents that have the same key words. Read the claims on those patents. What are other inventors claiming as their inventions? What are they not claiming? Then explain what claims you should be able to get, given what has already been claimed. Step 3: Evaluate the claims that you think you can get. What will they protect? What inventions will others be able to come up with that will get around your patent claims? Evaluate the nature of your patent claims. Are they broad or narrow? Are they enough to protect your product or service against a gainst imitation? Why or why not?

3. Looking at Patent Portfolios Go to www.chresearch. com for patent-based indicators indicators of 1,000 leading firms. Choose two firms in the same industry. Compare their patent portfolios. How are they similar? How are they different? Looking Looking at their patent portfolios, which company do you think has a better intellectual property position? Why?

1. Adapted from Jaffe, A., and J. Lerner. Lerner. 2004.

1 6004596.WKU.&OS PN/6004596&RS

Innovation and Its Discontents Princeton University Press. . Princeton, NJ: 2. http://patft.uspto.gov/ /patft.uspto.gov/netacgi/nphnetacgi/nphParser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p= 1&u=/netahtml/srchnum.htm&r=1&f=G&l=50&s

3. PN/6004596 Ibid. Ibi d. 4. McGavock, D. 2002. Intangible assets: A ticking time  bomb. Chief Executive http:/ http://findar /findarticles. ticles.com/p/ com/p/ articles/mi_m4070/is_2002_Nov/ai_94145235

199

 

Patents

5. Levin, R., A. Klevorick, R. Nelson, and S. Winter. Winter. 1987. Appropriating the returns from industrial research and development. Brookings Papers on

29. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 30. Kesan, Intellectual property property protection protection and agricultural biotechnology.

Economic Activity, 3: 783–832. 6. Mansfield, E. 1985. How rapidly does industrial technology leak out? Journal of Industrial Economics , 34(2): 217–223. 7. Levin, Klevorick, Nelson, and Winter, Winter, Appropriating the returns from industrial research and development. 8. Ibi Ibid. d. 9. Ibi Ibid. d. 10. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 11. Lemley Lemley,, M., and C. Shapiro. 2005. Probabilistic patents. Journal of Economic Perspectives, 19(2): 75–98. 12. Winter Winter,, S. 2000. Appropriating the gains from innovation. In G. Day and P. P. Schoemaker (eds.), Wharton on Managing Emerging Technologies Technologies . New York:  John Wiley Wiley..

31. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 32. Bercow Bercowitz, itz, L. 2000. Patent law changes: What you should know know.. Research Technology Management, 43(2): 5. 33. Fuerst, O., and U. Geiger. Geiger. 2003. From Concept to Wall Street: A Complete Guide to Entrepreneurship Entrepreneurship and Venture Capital. New York: Financial Times Prentice Hall. 34. U.S. Department of Commerce, Commerce, General Information Concerning Patents. 35. Flandez, R. 2005. Get a patent. Wall Street Journal, May 9: R9, R11. 36. U.S. Department of Commerce, Commerce, General Information Concerning Patents. 37. Ethe Etherton, rton, Let’s Talk Patents. 38. Montgomery Montgomery,, C. 2004. Drive-in dry-dock. The Plain

13. U.S. Department of of Commerce. 1992. General Information Concerning Patents. Washington, DC: U.S. Government Printing Office. 14. Kesan, J. 2000. Intellectual property protection protection and agricultural biotechnology biotechnology.. American Behavioral Scientist, 44(3): 464–503. 15. Ethe Etherton, rton, S. 2002. Let’s Talk Patents. Tempe, AZ: Rocket Science Press. 16. Yoffie, D. 2003. Intellectual property and a nd strategy, strategy,  Harvard Business School Note, Number 9-704-493. 17. Schil Schilling, ling, M. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill. 18. Yoffie, Intellectual property and strategy. 19. Munoz, S. 2005. Patent No. 6,004,596: Peanut butter and jelly sandwich. Wall Street Journal, April 5: B1, B9. 20. Yoffie, Intellectual property and strategy. 21. Jaffe, A., and J. Lerner. Lerner. 2004. Innovation and Our Discontents . Princeton, NJ: Princeton University Press. 22. Sandoval, G. 2005. Apple-Microso Apple-Microsoft ft duke out iPod fight in patent office. The Plain Dealer, August 17: C2. 23. Silverman, A. 1999. The forfeiture of U.S. patent rights by placing an invention on sale.  JOM, 51(2): 64. 24. U.S. Department of Commerce, Commerce, General Information Concerning Patents. 25. Ethe Etherton, rton, Let’s Talk Patents. 26. http://patft.uspto.gov/ /patft.uspto.gov/netacgi/nphnetacgi/nphParser?Sect1=PTO1&Sect2=HITOFF&d=PALL&p=1 &u=%2Fnetahtml%2FPTO%2Fsrchnum.htm&r=1&f  =G&l=50&s1=5,380,237.PN.&OS=PN/5,380,237&RS =PN/5,380,237 27. Ibi Ibid. d.

Dealer, July 20: C1, C6. 39. Ethe Etherton, rton, Let’s Talk Patents. 40. Regalado, A. 2004. Nanotechnology Nanotechnology patents surge as companies vie to stake claim. Wall Street Journal,  June 18: A1, A2. 41. Vara, V. V. 2006. Friendster patent on linking web friends could hurt rivals. Wall Street Journal, July 27: B1, B4. 42. Flandez, Get a patent. 43. Regalado, Nanotechnology Nanotechnology patents surge as companies vie to stake claim. 44. Ethe Etherton, rton, Let’s Talk Patents. 45. Allen Allen,, K. 2003. Bringing New Technology to Market . Upper Saddle River, NJ: Prentice Hall. 46. Rivette, K., and D. Kline. 2000. Discovering Discovering new value in intellectual property. property. Harvard Business Review, January–February: 2–10. 47. Kaminski, M. 2005. Effective Effective management of US patent litigation. Intellectual Property and a nd Technology Technology Law Journal, 18(1): 13–25. 48. Ethe Etherton, rton, Let’s Talk Patents. 49. Ibi Ibid. d. 50. Allen Allen,, Bringing New Technology to Market. 51. Jaffe and Lerner Lerner,, Innovation and Our Discontents. 52. Silverman, A. 2002. I’ll see you in court— Overview of a patent infringement trial.  JOM , 54(5): 64. 53. Ethe Etherton, rton, Let’s Talk Patents. 54. Gomes, L. 2005. Ebay wins fresh legal victory in challenge involving patents. Wall Street Journal, March 30: A6.

28. http://patft.uspto.gov/ /patft.uspto.gov/netacgi/nphnetacgi/nphParser?Sect2=PTO1&Sect2=HITOFF&p=1&u=%2Fn etahtml%2FPTO%2Fsearch bool.html&r=1&f=G&l=50&d=P  bool.html&r=1&f=G& l=50&d=PALL&RefSrch=ye ALL&RefSrch=yess &Query=PN%2F3841019

55. Greenhouse, L. 2007. Justices alter patent landscape. The Plain Dealer, January 10: C2. 56. Ethe Etherton, rton, Let’s Talk Patents. 57. Kaminski, Effective management management of US patent litigation.

200

 

Patents

58. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 59. Heinzel, M. 2005. BlackBerry maker agrees to settle patent dispute. Wall Street Journal, March 17: B5.

70. Reitzig, M. 2004. Strategic management of intellectual property. Sloan Management Review, 45(3): 35–40.

60. Bulkeley Bulkeley,, W. W. 2005. Patent ruling irks inventors, aids companies. Wall Street Journal, March 2: B1, B2. 61. Allen, Bringing New Technology Technology to Market. 62. Montgomery Montgomery,, Drive-in dry-dock. 63. Paris, E. 1999. David v. v. Goliath. Entrepreneur, November. 64. Varchaver archaver,, N. 2006. Who’s afraid of Nathan Myhrvold? Fortune, July 10, http://money.cnn.com/m http:/ /money.cnn.com/magaz agazines/f ines/fortune ortune/fortu /fortune_ ne_ archive/2006/07/10/8380798/index.htm. 65. Levy Levy,, S. 2006. The BlackBerry deal is patently absurd. Newsweek , March 13, http:/ htt p://www.msnbc.msn.com/ id/11677343/site/newsweek. 66. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 67. Ibi Ibid. d. 68. Ethe Etherton, rton, Let’s Talk Patents.

71. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 72. Ethe Etherton, rton, Let’s Talk Patents. 73. Richmond, R. 2005. Symantec patent may disturb rivals. Wall Street Journal, May 4: B3a. 74. Jaffe and Lerner Lerner,, Innovation and Its Discontents. 75. Ethe Etherton, rton, Let’s Talk Patents. 76. Levin, Klevorick, Nelson, and Winter, Winter, Appropriating the returns from industrial research and development. 77. Miller, S. S. 1996. Should patenting of of surgical procedures and other medical techniques be  banned? IDEA: The Journal of Law and Technology, 255–273. 78. Levin, Klevorick, Nelson, and Winter, Winter, Appropriating the returns from industrial research and development.

69. Chesbrough, H. 2001. The patent and licensing exchange: Enabling Enabling a global IP marketplace, Harvard Business School Teaching Note, Number 5-601-124.

79. Ibi Ibid. d. 80. Mullin Mullins, s, J. 2003. The New Business Road Test. London: Financial Times Prentice Hall.

201

 

202

 

Trade Secrets, Trademarks, and Copyrights

From Chapter 9 of Technology Strategy for Managers and Entrepreneurs Entrepreneurs.. Scott Shane. Copyright © 2009 by Pearson Prentice Hall. All rights reserved. 203

 

Trade Secrets, Trademarks, and Copyrights Learning Objectives Software Copyrights: Copyrights: A Vignette Introduction Secrecy When Does Secrecy Work? Trade Secrecy Secrecy as a Strategy Nondisclosure Agreements Noncompete Agreements Ownership of Intellectual Property Getting Down to Business: Using Nondisclosure and Noncompete Agreements Copyrights What Can Be Copyrighted? Who Gets a Copyright and How Do They Get It? Enforcement Through Litigation

Recent Developments to Strengthen Copyrights Software Copyrights Trademarks What Can Be Trademarked? Obtaining a Trademark  Enforcing a Trademark  Domain Names International Issues in Intellectual Property Differences in Intellectual Property Regimes International Agreements on Intellectual Property Discussion Questions Key Terms Putting Ideas into Practice Notes

Learning Objectives After reading this chapter, you should be able to: 1. Identify

the role that secrecy plays in protecting intellectual property. property.

2. Explain

when secrecy is an effective mechanism for deterring imitation.

3. Define

a trade secret and explain e xplain what characteristics are necessary for something to be a trade secret.

4. Explain

why nondisclosure agreements are an important part of efforts to maintain trade secrecy. secrecy.

204

 

Trade Secrets, Trademarks, and Copyrights

5. Define

a copyright, and explain how intellectual property can be protected by copyright.

6. Describe

how a copyright is obtained.

7. Define

a trademark, and explain why and a nd how trademarks are beneficial to companies.

8. Describe

how a trademark is obtained.

9. Explain

the major differences in intellectual property across countries, and their effect e ffect on technology strategy.

 Software Copyrights: A Vignette Vignette1  IBM learned a lesson about managing intellectual property the hard way. In 1980, when personal computer sales had reached $1 billion, IBM began to develop its first personal computer. The decision to develop a personal computer was an important and strategic one for the firm’s management, leading to much discussion of the best way to introduce the product. IBM’s senior management quickly concluded that time to market was important. So, to launch its product rapidly, IBM used off-the-shelf components, including Microsoft’s operating software, and Intel’s 8088 microchip. IBM’s management knew that the use of off-the-shelf components came at a risk. If the company used components available to others, then IBM’s computer would be easy to copy. But IBM’s management had a plan. The company would use a copyright to protect the proprietary code that its engineers were writing for the basic input-output commands that linked the computer hardware to the software. Unfortunately for IBM, this method of protecting the company’s intellectual property proved to be inadequate. A competitor then entering the personal computer business by the name of Compaq had its software programmers reverse engineer IBM’s input-output code. In a few months, Compaq was able to develop comparable code without infringing IBM’s copyright. How did Compaq do this? The answer lies in the type of intellectual property protection that copyrights provide. Like most software copyrights, the copyright on IBM’s input-output commands protected the code itself, but not the functions produced by the code. Consequently, as long as Compaq’s programmers used different software code to create the same functions as IBM’s code provided, they could copy the functionality of IBM’s input-output commands without violating IBM’s copyright. Compaq’s management thought up a really ingenious way to do this. First, it had a team of programmers reverse engineer the IBM software to figure out the functions that the code produced. Then the identified functions were given to a set of “virgin” programmers—programmers programmers—programmers who had never seen IBM’s code. These programmers were then told to write code for these functions. Because these programmers had never seen the IBM code, the code that they wrote was different from IBM’s. The end result was inputoutput software that provided identical functions to IBM’s but did not violate its copyright. This bit of ingenuity allowed Compaq to sell 47,000 IBM-

compatible personal computers in its first year, year, and set the company well on its way to becoming one of the most successful companies in the personal computer business. It also taught IBM’ IBM’s s management an important lesson about protecting intellectual property that you have now learned as well.

205

 

Trade Secrets, Trademarks, and Copyrights

I NTRODUCTION Protecting your (IP) is trademarks, more complicated than just learning how to company’s patent yourintellectual inventions.property Trade secrets, and copyrights are also important tools to protect intellectual property. For instance, Coca-Cola has made billions off of a beverage formula formula that is protected as a trade secret, Nike has turned its “swish” into a $7 billion trademark, and Disney derives much of its profits from animated characters and stories that it protects with copyrights. 2 This chapter discusses trade secrets, trademarks, and copyrights. The first section explains how you can deter imitation by keeping things secret and describes how companies use trade secrets as part of their technology strategies. The second section discusses copyrights and explains how you can use them to protect intellectual property and develop competitive advantage. The third section discusses trademarks and explains their role in technology strategy. The final section discusses the issues raised by differences in intellectual property protection across countries.

S ECRECY You can deter imitation by keeping things secret and reducing the leakage of information about your products or services or how you produce them (see Figure 1). For example, suppose that you have discovered a chemical that makes an excellent fertilizer. If you run a fertilizer company, you might not want other people to know that you have identified this chemical. If your competitors and potential competitors do not know that the key to your fertilizer lies in the use of a particular chemical, then

FIGURE 1

Protecting IP by Keeping Things Secret

You will run into problems if you talk too much about your intellectual property. property. Source: New Yorker Magazine. 1975. The New York Album of Drawings, 1925–1975. 1925–1975. New York: Penguin Books.

206

 

Trade Secrets, Trademarks, and Copyrights

they will not understand that they need to gain access to that chemical to compete with you successfully. Therefore, they will not seek to obtain access to that resource, and they will not be able to imitate your operations successfully. successfully.

When Does Secrecy Work? Efforts to mitigate imitation by keeping information about a new product or service secret work best under certain conditions. First, they work better when there are few sources of the information about the new product or service. To imitate your product or service, a competitor needs access to the information that makes copying the innoi nnovation possible. While your competitors can obtain this information from you, they can also get it from third parties. Your efforts to keep things secret are not going to be very effective if third parties readily provide this information to your competitors. Therefore, if only you know the information necessary to imitate your product or service, then your product or service is less likely to be copied. This is why it is easier for Coca-Cola to keep other companies from copying its soft drink formula than it is for your local dry cleaner to keep its dry cleaning formula secret. Even if your local dry cleaner never told anyone the formula for its dry cleaning solution, you could obtain it from any of thousands of other dry cleaners. However, if the few executives at Coca-Cola who know the formula to classic Coke do not tell you what it is, you are going to have no way of knowing it. Second, secrecy is more effective when a new product or service is complex. Imitation involves understanding how to copy a new product or service, not just having access to formulas or blueprints. The more complex a product or service is, the harder it is for people to figure out how to duplicate it. Complexity affects people’s understanding of the order in which tasks need to be undertaken and the difficulty of choreographing the joint efforts of different people. Take, for example, the difficulty of assembling a child’s toy. Even if you have the instructions, it is much harder to make the product just as the manufacturer had intended when the product is made up of hundreds of pieces than when it i t is made up of only a couple of pieces.3 Third, secrecy is more effective when the process of creating a new product or service is poorly understood. To imitate your activities, people have to understand what you are doing. The fewer competitors that can actually understand what you are doing, the fewer that will be capable of imitating your products, and the less imitation there will be. For example, suppose that you developed a new method for keeping storm drains clean by flushing them with a chemical mixture at certain temperatures. If the process of creating this new chemical solution sol ution was poorly understood (say that very precise amounts of the chemicals have to be combined at exactly the right moments under the right temperature for unknown reasons), then few people would be able to imitate this product, and your company comp any would capture the profits from providing it. Fourth, secrecy works best when the information that is being kept secret involves tacit knowledge—knowledge about how to do something that is not documented in written form. For instance, a plant manager’s knowledge of how to keep

an assembly line running at high speed spee d through a sense of where to position different di fferent workersthe with different skills and a salesperson’s knowledge of how to close sales by timing introduction of personal comments into a discussion are both examples of  tacit knowledge. It is easier to imitate a well-codified process than a tacitly understood one  becaus  bec ausee im imita ita tio n of a co codif difie ied d pro proces ces s on only ly req requir uires es acc acces esss to the do docum cument ent

207

 

Trade Secrets, Trademarks, and Copyrights

outlining the process, whereas imitation of a tacitly understood process requires the imitator to gain access to the person who holds that information in his or her head. Most of the time, it is easier to gain control of a document about a process than to gain control of a person who knows about it.4 Take, for example, the case of expertise in boiler repair. If that knowledge is held in documentary form by a company in Michigan, then a company in Ohio could get control of that information and move it to Ohio more easily than it could if the knowledge was tacit and held in the minds of the Michigan firm’s employees. To copy the tacit knowledge, competitors would need to hire the employees of the Michigan firm and get them to move to Ohio. Moreover, when knowledge is tacit, its transfer must take place through face-toface meetings between people. In contrast, when things are codified, knowledge can  be transferred by handing a blueprint or a formula to others. 5 Because knowledge spreads much faster if the transfer is not limited to direct contact between people, codified knowledge tends to spread very quickly, and is harder to keep secret than information that is not written down. Fifth, secrecy works better when there are limited numbers of people capable of understanding theabilities information being kept secret. The fewer who have the skills and to usethat theisinformation that creates yourpeople new product’s value, the fewer people that can figure out how to imitate what you are doing, even if the knowledge that you are keeping secret leaks out. Researchers Lynne Ly nne Zucker and Michael Darby at the UCLA business school have shown this to be true for new biotechnology companies. companies. They learned that the new biotechnology firms founded to exploit the technical expertise of leading scientists often were successful because competition was limited to the handful of people who also had the skills to exploit the cutting-edge scientific techniques that they used.6 Sixth, secrecy works better for processes, inputs, and materials than for products. Why? You sell your product in the marketplace. That makes the product itself  observable-in-use. (In fact, the more observable-in-use a product is, the less it can be kept secret. This is why it is hard to keep secret processes like techniques for providing customer service.7) Moreover, competitors can buy your product and reverse engineer it to figure out how it works. 8 These things make it harder for you to keep the composition of your product secret than it is for you to keep secret the production processes used to make the product. Therefore, processes make better secrets than products.

Trade Secrecy Trade secrecy is a special case of all efforts to keep a new product or service secret. In the United States, trade secrets are governed by state law, 9 primarily the Uniform Trade Secrets Act, which is in force in 44 states. 10 This Act defines a trade secret as “information “informatio n including a formula, pattern, compilation, program, device, method, technique, or process that derives independent economic value, actual or potential, from not being generally known, and not being readily ascertainable by proper means

 by,, other persons. . . Examples of trade  by trade secrets include: include: chemical chemical processes, customer customer databases, food recipes, computer source code, manufacturing processes, architectural designs, vendor lists, marketing plans, sources of raw materials, design manuals, 1 1 pricing policies, and blueprints. For instance, one of the most valuable trade secrets today is Google’s Web page ranking algorithm, which makes its search engine better than others.

208

 

Trade Secrets, Trademarks, and Copyrights

Trade secrecy laws provide for legal remedies if someone benefits from your trade secret without your consent. If you believe that someone else has improperly obtained your trade secret, you can sue to collect damages for your loss and obtain an injunction to stop further use of the secret. These remedies are available to you regardless of whether the party disclosing the trade secret was bound by a duty of  confidentiality, had signed a nondisclosure agreement, obtained the information illegally,, obtained the information from someone who did not have authorization to disgally close it, or learned the information by accident, but knew it was a trade secret. 12

Conditions to Have a Trade Secret  Three conditions must be met for the courts to hold that something is a trade secret. First, the information must be known only by people in your company. Information that is known generally in an industry industry,, such as standard manufacturing processes, or information that can be generated from data that are known in an industry, cannot be a trade secret.13 Moreover, you cannot claim that the general skills that your employees learn on the job are trade secrets because that would preclude them from being able to take new jobs and use the skills that they learned working for you at their new employers.14 Second, the information must have economic value. For something to be a trade secret, it must generate a competitive advantage that would be lost if your competitors made use of it. This means that you must be able to document that what you term a trade secret is important to how your company derives value, and provides an advantage over your competition in the market place. You should note that this standard is stricter than for a patent, where all you have to do is prove infringement to collect damages. Third, you must take reasonable measures to keep the information secret. This means that you have to adopt “secrecy policies” to ensure that people do not accidentally access the secret information. Your employees need to know what information is secret and that secret information is limited to only those personnel who need it. Moreover, those personnel who need access to the information must agree, in writing, to keep it confidential. Furthermore, you need to use physical mechanisms, such as limitingpasswords the access and of nonemployees your facilities, from locking files, requiring computer so on, to keeptothe information 15 getting out. Take, for example, the efforts by KFC to keep the recipe for its fried chicken a trade secret. The recipe is kept in a vault at the company’s headquarters, and only a few people know what it is. Those T hose employees who know the formula are required by the terms of their employment to keep the recipe secret. Moreover, two different companies supply the herbs and spices to KFC, but each one is allowed to create only part of the ingredients, and neither company is known to the other.16

Secrecy as a Strategy You might choose secrecy as your basic approach to protecting intellectual prop-

erty. This choice may stem from a preference for trade secrets over patents as the  basis formutually competiti compe titive ve advant advantage age (the two asources intellectua intel lectual l prope property rty secrecy protec-protec tion are exclusive, necessitating choice),ofperhaps because trade offers a longer time horizon of protection or because it does not disclose information to your competitors. Or it may occur because you have a product for which secrecy is particularly effective: It is created through a process that is poorly

209

 

Trade Secrets, Trademarks, and Copyrights

understood, complex, and based on tacit knowledge for which there are few sources of information and a limited number of people who can comprehend it. You might even focus on secrecy to generate customer interest in your products and services because people are often more interested in things that they can’t know about than things that they can. Apple Computer is an example of a high-technology company that focuses very much on secrecy. secrecy. (A former CEO, John Scully, Scully, was fond of using the phrase, “loose lips sink ships.”) The company rarely discloses its plans for new products and compartmentalizes development efforts so that employees working on new products rarely have information about the entire product. The company vigorously maintains efforts to limit disclosure, suing employees that leak information about forthcoming products and Web sites that publish such information. It creates lists of  employees that have been given access to information about new product plans, even watermarking documents with the recipient’s name, and using different code numbers for different departments to better track the source of any leak. Access to  buildings in Apple’s headquarters is even limited to the part of the complex in which employees work.17 While secrecy-focused strategies, as Apple’s, have many advantages, these benefits come at a cost. As was such mentioned previously, maintaining trade secrets requires the adoption of secrecy policies and reduces the level of informal exchange of information among your employees, which hinders your ability to develop new products and processes. Maintaining trade secrets also inhibits your efforts to work with other companies, which, by necessity, lack adequate information to serve as effective partners. Moreover, it hinders efforts to sell products to many business customers, who need to know kno w about new products long in advance of their release to fit them into their own plans. Finally, maintaining trade secrets risks the independent discovery and exploitation of your inventions. Competitors who independently and legally obtain technology that you maintain as a trade secret—e.g., by reading your publications, talking to your suppliers or customers, and reverse engineering your products—are free to use it to make and sell exactly the same products as you, even though they would be barred from doing so if you patented the technology. technology. (Table (Table 1 describes the conditions when protecting intellectual property with trade secrets is better than protecting it with patents and vice versa.)

TABLE 1

Trade Secrets Versus Patents This table shows when it is better to protect an invention with trade secrecy and when it is better to patent an invention. TRADE SECRE ECRETS TS ARE BETTER WHEN . . .

PAT ATENTS ENTS ARE BETTER WHEN . . .

The secret cannot be patented

Reverse engineering is possible

The product cycle is short The patent would be hard to enforce

The market life of a product is close to 20 years The patent is easy to enforce

The patent would be narrow

The patent would be broad

It is hard to identify a trade secret

It is easy to identify a trade secret

2005. Marketing Source: Adapted from Mohr, J., S. Sengupta, and S. Slater. 2005. Market ing of High-T High-Techno echnology logy Produ Products cts and and Innovati Innovations ons (2nd edition). Upper Saddle River, NJ: Prentice Hall.

210

 

Trade Secrets, Trademarks, and Copyrights

Nondisclosure Agreements As Figure 2 shows, trade secrecy is enhanced by having having people sign nondisclosure agreements that are crafted by lawyers who know the details of employment law. These agreements are important; you cannot make a case that you are keeping information secret unless your employees understand that they are expected to refrain from disclosing information. Effective nondisclosure agreements must meet certain conditions. The agreements must specify exactly what information is to be kept secret, and cannot state that all information that employees learn during their employment is confidential. Moreover, the agreement must provide consideration . That is, employees must receive something of value, like their salaries, in return for nondisclosure. Furthermore, the agreement must specify legitimate uses for the information, including identifying those people to whom the information can be disclosed, and how the information may be used to perform a job. Lastly, the agreement must state what must be done with any documents or materials that are transferred to the employee, both during employment and after the termination of an employment relationsh relationship. ip.18

Enforcing Nondisclosure Agreements  To enforce nondisclosure agreements, you need to be willing to sue your employees and others who help them because the only remedies for violation of nondisclosure agreements come through legal action. Many companies do this. For example, Biomec Inc, a Cleveland, Ohio, medical device company, sued a former employee claiming that he violated his confidentiality agreement when he moved to rival, Cleveland Medical Devices; and Wal-Mart sued Drugstore.com and the venture capital firm Kleiner Perkins when Drugstore.com hired former Wal-Mart employees who had developed that company’s system for Internet retailing. 19

FIGURE 2

Nondisclosure Agreements Help to Protect IP

Companies often protect their intellectual property by having people sign nondisclosure agreements that preclude them from letting anyone else know the protected information. Source: United Features Syndicate Inc., March 16, 1996.

211

 

Trade Secrets, Trademarks, and Copyrights

While the easiest case to make for violation of a nondisclosure agreement occurs when your employees take documents do cuments that belong to your company comp any,, you can make a case that they violated their nondisclosure agreements if they take only uncodified knowledge. For instance, IBM recently settled a lawsuit with Compuware Corp. in which Compuware alleged that IBM had violated Compuware’s trade secrets for file management and error detection software by hiring former Compuware employees to speed the development of software for its mainframe computers. In this case, Compuware claimed that its former employees had signed confidentiality agreements and then disclosed technical knowledge and knowledge of customer preferences to IBM.20

Noncompete Agreements Trade secrecy is enhanced by having your employees sign noncompete agreements, which bar them from working for competitors for a period peri od of time after their employemplo yment has ended, because these agreements keep employees from moving to rivals while their company-specific knowledge still has value (see Figure 3 for an example of a nondisclosure and noncompete agreement). For example, Microsoft successfully forced a start-up company named CrossGain to lay off 20 former Microsoft employees until the expiration of their noncompete agreements, as a way to protect its intellectual property.21

Enforcing Noncompete Agreements  As with nondisclosure agreements, you need to be prepared to go to court to enforce your noncompete agreements. For example, Patio Enclosures Inc. had to take Four Seasons Solar Products to court for hiring a former Patio Enclosures employee who had signed a noncompete agreement that barred him from employment at a competing firm for two years.22 While noncompete agreements help you to protect your company’s intellectual property,, they are hard to enforce. These agreements need to be of limited length and property limited geographic breadth because they will be declared invalid if they keep people from earning a living in their chosen field.23 (For example, ExxonMobil’s noncompete agreement cannot preclude a petroleum engineer from working at another oil company after leaving ExxonMobil.) Moreover, in many states, you must give employees some benefit, like a bonus or a higher salary, in return for asking them to sign a noncompete agreement;24 in other states, like California, you cannot enforce these agreements at all.25

Ownership of Intellectual Property Related to the issue of nondisclosure and noncompete agreements is the issue of  who owns the rights to technologies that employees develop during the period of  their employment at a company. These rights reside with employees unless you require them to assign the rights to you. Of course, most large companies do just

this, which keeps many people from quitting and starting new companies to exploit technologies that they developed while working elsewhere. For  Jeff Hawkins, the founder of Palm computing, computi ng, patent patented ed an algorithm algori thm forinstance, pattern patter n recognition software when he was on academic leave from GRiD systems, his employer.. Although he owned the patent to the algorithm, and his licensing agreeemployer ment with GRiD allowed him to use it in noncompeting products, he did not have

212

 

Trade Secrets, Trademarks, and Copyrights

FIGURE 3

Nondisclosure/  Noncompete Agreement

This agreement is made as of the 23rd day of June, 2002, by and between: ACME Inc.  Inc.  located in CITY CI TY,, STATE STATE and JOHN INVENTOR located INVENTOR located in CITY, STATE. This agreement shall govern the conditions of disclosure by JOHN INVENTOR to INVENTOR to ACME Inc.  Inc.  of certain “Confidential Information” including but not limited to prototypes, drawings, data, trade secrets and intellectual property relating to the “ Patent Pending ” invention named “Mouse Trap” Trap” invented by JOHN INVENTOR. INVENTOR. With regard to the Confidential Information, ACME Inc. hereby Inc. hereby agrees: 1. Not 1.  Not to use the information therein except for evaluating evaluating its interest in entering a business relationship with JOHN INVENTOR, INVENTOR, based on the invention. 2. T 2.  To o safeguard the information against disclosure to others with the same degree of care as exercised exercis ed with its own information of a similar nature. 3. Not 3.  Not to disclose the information to others, without the express written permission of JOHN INVENTOR,, except that: INVENTOR a. which ACME Inc. can Inc. can demonstrate by written records was previously known; b. which are now, now, or become in the future, public knowledge other than through acts or omissions of ACME Inc.; Inc.; c. which are lawfully obtained by ACME Inc. from Inc. from sources independent of JOHN INVENTOR; INVENTOR; 4. That 4.  That ACME Inc. shall Inc. shall not directly or indirectly acquire any interest in, or design, create, manufacture, sell, or otherwise deal with any item or product, containing, based upon or manufacture, derived from the information, except as may be expressly agreed to in writing by JOHN INVENTOR.. INVENTOR 5. That 5.  That the secrecy obligations of ACME Inc. with Inc. with respect to the information shall continue for a period three years from the date hereof. JOHN INVENTOR will INVENTOR will be entitled to obtain an injunction to prevent threatened threatened or continued violation of this Agreement, but failure to enforce the Agreement will not be deemed a waiver of this Agreement. IN WITNESS WHEREOF the Parties have hereunto executed this Agreement as of the day and year first above written. ACME Inc. By:

Date:

Title:

JOHN INVENTOR and INVENTOR and SIGNATURE

This figure provides an example of a nondisclosure and noncompete agreement. Source: Inventnet, http://www.inventnet.com/nondisclosure.html.

the rights to improvements to the C-language enhancements he had made while a GRiD employee. As a result, he needed to work around this intellectual property to develop the Palm personal digital assistant.26

 Key Points • Keeping things secret is an important way to protect your intellectual property. • Secrecy is most effective when there are few sources of information about a product, when information about it is complex, when the process of creating the

213

 

Trade Secrets, Trademarks, and Copyrights

GETTING DOWN TO BUSINESS

Using Nondisclosure and Noncompete Agreements

A product development development engineer who was working for a large medical device company in Minnesota had developed a new product to replace existing cardiac pacemakers. Because several aspects of the product were novel, nonobvious, and useful, it was eligible eli gible for patent protection. Moreover, the product was much  better than exist existing ing cardia cardiacc pacema p acemakers, kers, and all the cardiologists who had seen a prototype were very enthusiastic about it. While the previous paragraph might seem like the introduction to a tale about a technology entrepreneur who becomes fabulously wealthy, that is not the case here. You see, the engineer had signed nondisclosure and noncompete agreements as conditions of his employment at the medical device firm. Like most nondisclosure agreements, the one that this employee had signed precluded him from disclosing proprietary information that was developed during his employment at the company. And like most noncompete agreements, this one also barred him from working for a competitor for several years after leaving the medical device firm. Even though the medical device firm was not interested in pursuing the new product and did not even want to patent it, the engineer couldn’t start a company or work with another company to bring the new product to market because of the noncompete

•

• •

•

and nondisclosure agreements that he had signed. The medical device firm that owned the intellectual property behind the cardiac pacemaker could take to court him and anyone else who hired or helped him. The threat of litigation was enough to stop the engineer from starting a company. He knew that it would be foolish to start a company that would immei mmediately be enmeshed in a lawsuit that would take a large amount of his time and would require him to pay large legal fees that he would have no revenues to cover. Moreover, he could not even take a job at another company working on the development of similar products. While other medical device companies were eager to hire him, believing that he would develop valuable new products, they were afraid to hire him to work on cardiac products because his former employer could then sue them, claiming that he had disclosed its knowledge to competitors. The lesson here is that companies can often use nondisclosure and noncompete agreements to prevent their employees from using the intellectual property developed during their period of employment. This is why the creation and enforcement of strong nondisclosure and noncompete agreements are important parts of a technology strategy.

product is poorly understood, when the knowledge necessary to create it is tacit, when few people can understand that knowledge, and when value comes from a process rather than a product. Trade secrecy is a special case of efforts to keep information secret; to have a trade secret, a piece of intellectual property must be of value, must not be known generally, generally, and must be kept secret. To protect a trade secret, you must take action to keep it secret by such activities as limiting access to facilities and having employees sign nondisclosure agreements. Nondisclosure agreements are legal documents in which employees agree to keep a company’s information secret; noncompete agreements are legal documents that preclude employees from working for competitors for a period of  time after their employment has ended. Noncompete and nondisclosure agreements are difficult to enforce because

enforcement requires legal action, because companies cannot prevent people from earning a living inand their chosenemployees fields, because agreements must bein precise to be effective, because must the be given consideration return for signing the agreements. • Companies typically require their employees to assign them the rights to intellectual property developed during their period of employment.

214

 

Trade Secrets, Trademarks, and Copyrights

C OPYRIGHTS copyright A is27a It legal protection to the authors of or original literary, musical, or artistic works. gives the rightgiven to reproduce, display, produce derivative works from the protected item. It also gives the right to sue to collect damages if someone else infringes the copyright from the time the work was created until 70 years after the author’s death (or 95 years after publication for works for hire—more about that next). Infringement occurs if another party duplicates, displays, produces, or distributes the work, or gives, rents, or lends it to others.

What Can Be Copyrighted? Many things can be protected by copyright, including books, movies, software, music, other recordings, databases, plays, pantomimes, dances, sculptures, graphics, and architectural designs.28 For example, the LEGO Group has used copyrights to protect the appearance of its standard, eight-studded block against imitation.29 The thing that you want to protect does not need to be novel or even lawful to receive copyright protection. For instance, Napster’s software copyright still holds even after the file-sharing service was deemed unlawful. 30 However, there are some limitations on what can be protected by copyright. First, the thing being protected has to be tangible. Thus, you cannot copyright an impromptu speech, but you can copyright a written one. Second, titles and names cannot be protected by copyright; these things are protected by trademarks. Third, slogans, ideas, methods, principles, discoveries, or things composed of common property, like calendars, cannot be copyrighted. 31 Fourth, the work has to be produced because the idea behind it cannot be copyrighted.

Who Gets a Copyright and How Do They Get It? So who can get a copyright? A copyright can be obtained obtained by the author of any completed original work, unless the work is done for hire. Work for hire is a technical term for work that is done under the scope of a person’s employment, or under a written agreement between the author and the person contracting for the work, which requests that the work be done on the contractor’s behalf. If the work is done for hire, then the copyright goes to the entity commissioning the work. For instance, you could hire your roommate to write some software for the insurance claims adjusting business that you are starting, and the copyright on that software would then belong to you. 32 You can obtain a copyright without taking any action other than putting the intellectual property into tangible form (for instance, writing something on paper or recording it on a DVD or CD). Alternatively, as Figure 4 shows, you can apply for copyright protection from the U.S. Patent and Trademark Office. While applying for a copyright is not necessary, it does provide a couple of important advantages. Most notably, you need to have a registered copyright to file a lawsuit to protect your copyrighted intellectual property; so registration is useful in the event that you want

to sue someone.

Enforcement Through Litigation If you think that someone has improperly used your copyrighted materials, you can take them to court and sue them for infringement. Because plaintiffs in a

215

 

Trade Secrets, Trademarks, and Copyrights

copyright infringement lawsuit rarely have direct evidence of the actual incidence of copying (of course, having photos of people in the act of copying your copyrighted material and distributing it would strengthen your case!), the courts usually infer that copying has occurred if the new work is substantially similar to the copyrighted work, and the defendant had access to it. If you win a copyright infringement lawsuit, the court will award you damages. The size of those damages depends on the intent of the infringer, how much money they made, how they made their money, and how their actions affected your business. For instance, if the infringer charged others for your copyrighted material, then the size of the damages that you can receive will be greater than if  they gave away your material for free. Also, the size of the damage settlement will

FIGURE 4

A Copyright Application Form

(continued)

216

 

Trade Secrets, Trademarks, and Copyrights

This figure shows a U.S. copyright application form for a copyright on a visual art. http://www.copyrig /www.copyright.gov ht.gov/forms /forms/formv /formva.pdf. a.pdf. Source: http:/

 be larger if the infrin infringer ger reduces the commerci commercial al value of your propert property y through their actions. If the court determines that the infringer’s imitation was intentional, then you can collect triple the value of your loss as damages. So it is important to affix the copyright symbol (©) to your material. Doing so allows the court to reject any claim  by an infringer that he or she did not know the material was copyrighted and innoi nno33 cently infringed.

If you believe that someone has infringed your copyright, you can ask the court to issue an injunction, stopping that party from using your copyrighted material while the case is being decided. However, if you believe that your copyrighted material has been infringed, you need to take action quickly. The statute of limitations on copyrights only lasts three years.

217

 

Trade Secrets, Trademarks, and Copyrights

Recent Developments to Strengthen Copyrights Although copyrights were originally intended to protect written documents, in recent years, most of their growth has been as a means of protecting sound and images (as well as computer software). Now such things as video recordings of  Super Bowl games and Web casts of the weather outside of college dorms are routinely copyrighted. However, software is easier to copy than books and other printed material  because  becau se dupl duplicat icating ing and dist distribu ributing ting multi multiple ple copi copies es of a book take takess more time and money than duplicating and distributing a piece of software. Therefore, copyright violations have been increasing in the digital age. To deter copying and illegal distribution, software companies often impose very restrictive end user license agreements (EULAs). By severely limiting how their customers can use their products, these companies strengthen their position against violators of their copyrights. Unfortunately for copyright holders, the development of computer network technology to share digital files has made it easier to copy protected material, reducing sales of the legitimately duplicated versions. For instance, file sharing technology has made it easy to copy musical recordings, leading to declining sales of music CDs. As a result, copyright holders have become more vigilant about protecting their intellectual property. property. In the case of digital music files, the record labels have begun to sue anyone that does anything that lets users get around their copyrights. For example, several record labels recently sued XM Satellite Radio because XM’s Inno device allows users to record, store, and create play lists of songs that they have heard on XM. The record labels claim that the use of the Inno device violates the copyrights to their songs by allowing people to obtain recordings of them without paying a royalty.34 Similarly, the record labels sued Napster for making it possible for people to exchange digital music files without paying royalties, charging that the company violated the copyrights of recording artists, and caused them financial loss. The record labels’ argument was that Napster created a market in which other people could avoid paying royalties on copyrighted songs, thus enabling infringement.35 (The copyright issue wasn’t settled in this case because Napster was forced to shut down when the judge in the case issued an injunction banning Napster from offering the service until the courts had decided the case.36 However, many observers  believe that file-sharing networks will not be able to claim “fair use” of copyrighted material—see Figure 5.) While the recording industry was able to use the court system to enforce its copyrights against the first generation of peer-to-peer networks like Napster, they face a more difficult time with second generation peer-to-peer networks that don’t use a central server for file sharing. The use of second generation peer-topeer networks spreads the copyright violation across numerous parties and makes the value of their infringement too small to justify the use of lawsuits as a way to stop it. 37 Recently, laws have been enacted to let companies use physical tools, such as

embedded authentication chips, to make it more difficult for people to copy a piece of intellectual property. For instance, the Audio Home Recording Act of 1992 requires that all digital recording devices include a Serial Copy Management System, which permits originals, but not copies, to be duplicated. And the Digital Millennium Copyright Act (DMCA) made it illegal to circumvent a technological device that is used to prevent duplication of copyrighted material. 38

218

 

Trade Secrets, Trademarks, and Copyrights

FIGURE 5

Fair Use

The fair use of a copyrighted work, including such use by reproduction in copies or phonorecords phonorecor ds or by any other means specified by that section, for purposes such as criticism, comment, news reporting, teaching (including multiple copies for classroom use), scholarship, or research, is not an infringement of copyright. In determining whether the use of a work in any particular case is a fair use the factors to be considered shall include— 1. The purpose and character of the use, including whether such is of a commercial nature or for nonprofit educational purposes; 2. The nature of the copyrighted work; 3. The whole amount and substantiality of the portion used in relation to the copyrighted work as a whole; and 4. The effect of the use upon the potential market or value of the copyrighted work. The fact that a work is unpublished shall not itself bar a finding of fair use if such a finding is made upon consideration of all of the above factors.

The U.S. government provides a “fair use” exemption to copyright law, which allows a user to make use of copyrighted material under the conditions listed above. Source: Section 107 of the Copyright Act of o f 1976, http://www.copy /www.copyright.gov/title17/92chap1.html#107. right.gov/title17/92chap1.html#107.

However, the use of copy protection software to prevent sharing of intellectual property has had problematic side effects. It limits the devices that customers can use to play legitimately purchased recordings and sometimes causes damage to computers that play the recordings.39 For instance, Sony BMG recently had to reimburse its customers more than $100 each for computers damaged by hidden antipiracy software that Sony BMG had placed on their CDs.40 Efforts to strengthen copyrights have had other adverse effects as well. They have diluted the concept of “fair use” of copyrighted material. As a result, it is  becoming  becom ing more diffi difficult cult to make nonco noncommerc mmercial ial use of thes thesee mate material rials. s. Secon Second, d, these efforts have hindered the natural process by which innovators build on the work of others by requiring them to obtain the rights to use any copyrighted material simply to build on it.

Software Copyrights Copyrights have become an important mechanism to protect software. While the mathematical formulas and equations underlying software programs are not copyrightable, nor are the ideas or methods behind them, copyrights can be used to protect many parts of computer software, including source code, object code, microcode, and screen displays.41 For instance, ConnectU.com, a social networking Web site, uses copyrights to protect its source code. 42 While copyrights provide some intellectual property protection for software,

they are not an ideal form of protection for this medium because they only protect the expression of ideas, not the concepts underlying those ideas. As the introductory vignette in the chapter pointed out, ideas can often be expressed in a variety of different ways, allowing someone to reverse engineer a piece of software and then write a new piece of software that works around the copyright  by ex expre pre ss ssin ing g th thee sa same me id idea ea in a di diffe fferen ren t way. If a de defe fend ndan antt in a so softw ftw are

219

 

Trade Secrets, Trademarks, and Copyrights

copyright case can show that they created a work independently and expressed an idea in a different way from the holder of the copyright, then there is no copyright violation.43 Moreover, demonstrating the infringement of a software copyright is not easy to do directly, making software copyrights difficult to enforce. Because it is impossible to show an exact linkage between the expression of an idea and the process of  expressing it, courts have had to interpret the “look and feel” of software to determine whether copyright infringement has occurred. Of course, this reliance on “look and feel” to determine infringement makes it harder to know if infringement has actually occurred.44 On the other hand, software copyrights provide additional intellectual property protection to that that provided by software patents. A wider variety of software programs can be copyrighted than patented because any originally authored work presented in tangible form can be protected by a copyright, while only novel, nonobvious, and useful inventions can be protected by a patent. Copyrights also are much easier to obtain than patents, and are, consequently, a much less expensive form of protection. Furthermore, copyrights offer protection until 70 years after the author’s death, while patents offer offer protection for only 20 years after after the time of  invention. 45  Key Points

• Copyrights give the authors of original works the right to distribute, duplicate, and provide derivations of that work, and to preclude others from doing the same. • A variety of things can be copyrighted, including literary works, dramatic works, audio and video recordings, and computer software; however, intangi ble things, titles, names, slogans, ideas, methods, principles, princip les, and works composed of common property p roperty,, cannot be copyrighted. • Copyrights are given to the author of any original work, unless the work is done for hire. • Copyrights can be obtained either by putting the work into tangible form or by registering the work at the USPTO; registration provides the right to sue for copyright infringement. • Copyrights offer a negative right; they do not stop others from copying your intellectual property; they only give you the right to sue to obtain damages if  your copyrighted material has been infringed. • File-sharing software poses an important threat to copyrights on recorded music, and its rise has led to a number of infringement lawsuits. • Recent laws have strengthened the position of copyright holders by allowing them to use physical tools to prevent duplication of their work; however, these physical tools have had problematic side effects. • Copyrights can be used to protect the source code, object code, microcode, and screen displays in software but not the ideas, mathematical formulas, or equations behind them.

• Because it is impossible to show the exact link between the expression of an idea the process underlying it, courts interpret the “look and feel’ of softwareand to evaluate infringement. • Copyrights are less effective than patents at protecting software but can be obtained to protect a wider range of things.

220

 

Trade Secrets, Trademarks, and Copyrights

T RADEMARKS 46

Trademarks are devices to identify theprotection provider ofthan a product or service. While they offer much less intellectual property patents, copyrights, or trade secrets, they do help companies to protect their brand names. For instance, the Intel Inside® trademark helps Intel build its brand by making it easier for that company to differentiate itself from competitors. In addition, trade and service marks can be used as leverage to drive other forms of strategic advantage. For instance, Cisco recently settled a lawsuit with Apple Computer over violation of its iPhone trademark. Cisco wanted Apple to make its iPod and iPhone products compatible with non-Apple products. By blocking Apple’s use of the iPhone name, Cisco forced Apple to concede on the issue of  compatibility.47 Because consumers associate particular trade or service marks with the quality of the products or services that companies provide, some trademarks are quite valuable. For instance, the Microsoft trademark is now worth $60 billion. 48 Therefore, learning how trademarks work and how they protect intellectual property is an important part of o f technology strategy.

What Can Be Trademarked? A trade or service mark can be obtained on any word, number, number, symbol, phrase, color, design, or even smell that distinguishes the products and services of one company from those of another. For instance, Nike has trademarked its “swoosh” symbol, while Porsche AG has trademarked the numerical sequence “911.”49 However, not everything can be trade or service marked. For a word, number, symbol, phrase, color, design, or smell to be appropriate as a mark, it cannot describe des cribe the product or service that a company provides. For instance, a supermarket cannot trademark the word carrot carrot because  because that word is descriptive of the products sold at a supermarket. However, an airline could trademark that word because carrots are not descriptive of what airlines do. A common word, word, like house cannot be trademarked. However, what is a common word depends on interpretation interpretation by the courts. A Federal appeals court recently recently entrepreneur,, allowing that upheld Entrepreneur Magazine’s trademark on the word entrepreneur company to block the use of that word by others. Ironically, the fact that someone else has trademarked a word, number, sym bol, phras phrase, e, color, co lor, design, desi gn, or o r smell sme ll does d oes not n ot mean me an that tha t you can’t use u se the th e same sam e one. one . A trade or service mark can be be used by more than one company company if customers would not be confused about the identity of the provider of the product or what the product is used for, and if the use by a second party does not dilute the value of the mark. Typically, this means that a mark can be used by two companies if  they sell different types of products and services (e.g., airplanes and vegetables) through different channels. For instance, Apple Computer and Apple Records are  both able to have h ave a trade trademark mark with the word w ord apple in it because personal comput-

ers and Beatles songs are very different products and are sold through different marketing Apple moves further music  bus  busine ine ss ss,, itchannels. may ma y fac faceeHowever, proble pro ble msasusi ng itsComputer tra dem ark ed nam e for into tha tthe busine bus ine ss  because its name n ame might then cause confus confusion ion among custom customers ers as to the provid provider er of the product.

221

 

Trade Secrets, Trademarks, and Copyrights

FIGURE 6

Many Companies Obtain a Large Number of  Trademarks

A Mathematical Assistant™ ANYLITE™

MBA™ muLisp™

AOS™ APD™ Automatic Power Down™ Avigo™ BA II PLUS™ BA Real Estate™ Calculator-Based Laboratory™ CBL™ CBL 2™ Calculator-Based Ranger™ CBR™ CellSheet™ Clear Calc™ Constant Memory™ Datamath™ Data Synchronization™ Derive™ Derive™ 5 DockMate™ EOS™ Europa™ Executive Executiv e Business Analyst™ EXPLORATIONS™ Explorer™ Explorer Plus™ Financial Invest Investment ment Analyst™ five2eight™ Galaxy™ GeoMaster™ Home Manager™ Hot Calc™ LearningCheck™ Math Explorer™ MathMate™ Math Star™

NoteFolio™ Paper-Free™ Paperless Printer™ Personal Banker™ Phone Bank™ Pocket Dialer™ PocketMate™ Pocket Paper-Free Printer™ Pocket Speller™ Pocket Thesaurus™ Pro-Calc™ Profit Manager™ Student Business Analyst™ Study Cards™ SuperBundle™ SuperView™ T3™ Teachers Teaching with Technology™ T3 Europe Talk TI™ Technofiscaphobia™ Tfas™ TI-30Xa School Edition™ TI-30Xa SE™ TI-83 Plus™ TI-Cares™ TI Connect™ TI FLASH Studio™ TI-GRAPH LINK™ TI InterActive InterActive!™ !™ TimeSpan™ TI-Navigator™ TI-Presenter™ TI-TestGuard™ ViewScreen™ Voyage™ 200

This figure shows a partial list li st of trade and service marks belonging to Texas Instruments. Instruments. Source: Adapted from http://education.ti.com/us/global/trdmrk.html.

Because trade and service marks are valuable tools, many large, established companies have obtained a large number number of marks. Figure 6 shows just a partial list of the trademarks obtained by one technology company—Texas Instruments.

Obtaining a Trademark So how do you get a trademark? In common law countries, like the United States, you get a trade or service mark by using the word, phrase, symbol, design, or smell

or by registering that mark with the USPTO.50 The process of registering a trade or service mark is very simple. You just send an application to the USPTO along with a drawing of the mark and the payment of the fee for the relevant category of mark. However, before you send in your application and pay the money to register a trade or service mark, you probably want to conduct a trademark search. The USPTO

222

 

Trade Secrets, Trademarks, and Copyrights

is not going to give you a mark that violates that of another company. Conducting a search will minimize the chances that you will select something that infringes on another mark, as well as the likelihood that you will select something that cannot be trademarked. Although you will not get a trade or service mark right from the USPTO until you use a mark, and you don’t need to register the mark to enforce it, you probably want to go down the registration route when you seek trademark protection. Registration provides a record of your claim of ownership of the mark, which is useful to signal your actions to competitors. In addition, you cannot sue to protect your trade or service mark, or collect triple damages in the case of infringement, until the mark has been registered.51 Furthermore, registration makes it easier to obtain trade or service mark rights in other countries. 52

Enforcing a Trademark Once you have registered a trade or service mark, your ownership of it lasts for ten years can be renewed longyou as have the mark is in the usemark, and has 53 However, dated.and five yearsasafter obtained younot willbeen needinvalito file an affidavit with the USPTO attesting that the mark is still in use. If you don’t do this, your trade or service mark can be cancelled. Trade and service marks can be invalidated by the USPTO in one of three ways: through cancellation proceedings, through abandonment, or through through generic meaning. Cancellation occurs when the owner of the mark fails to attest to its continued use. Abandonment occurs when someone else can show that the owner of the mark has stopped using it. For example, in the recent dispute between Cisco and Apple Computer over the trademark “iPhone,” Apple Computer sought to show that Cisco hadn’t sold iPhone-branded products for a period of time, thus indicating that Cisco had abandoned the trademark.54 The potential for abandonment is why trademark holders fight hard to protect their trademarks. For instance, Entrepreneur Magazine Incorporated fights to exert its rights to the trademark entrepreneur against a variety of small companies, not  because it thinks it would obtain any significant royalties from enforcing the trademark, but to defend the use of the mark against other companies that claim that Entrepreneur Magazine has abandoned it. Generic use occurs when a mark no longer represents a specific product or service and ends up representing a general category of products or services (as occurred, for example, with the once trademarked term escalator escalator). ).55 Once a trade or service mark becomes a generic term, it reverts to the public domain and anyone can use it. That is why Bayer works hard to ensure that Aspirin® is not used to refer to all pain medications. medications. If that were to occur occur,, the word could no longer longer be trademarked because it would no longer distinguish Bayer’s product from those of other companies. Like other forms of intellectual property protection, trade and service marks are

enforced through legal action. Owners of a mark can sue to prevent both infringement and dilution of the value of the mark. 56 Infringement occurs when a competitor’s use of a mark causes confusion amongst customers about the provider of a product. For example, VoIP start-up, Vonage, Vonage, has sued AT&T claiming that the name of AT&T’s AT&T’s VoIP VoIP service, CallVantage, violates its trademark because that name is too close to its own.

223

 

Trade Secrets, Trademarks, and Copyrights

Dilution occurs when another party’s use of a word, phrase, symbol, design, or smell lowers the value of a company’s trade or service mark. For instance, American 57 Express able to that stopitsa trademark’s limousine service from using by thethat name “American Express” was by showing value was reduced action. You need to protect your trademarks. Failure to take legal action to enforce your rights can result in the loss of a trade or service mark through abandonment. Unfortunately, taking legal action costs money; and many organizations fail to protect valuable trademarks. For instance, the Metropolitan Transit Authority (MTA) in New York City has trademarked its circular route symbols for the A, D, F, 1, 4, and 7 trains. However, many companies frequently violate the MTA’s MTA’s trademarks by making unauthorized T-shirts—or, T-shirts—or, in the case of Eli Zabar ’s food emporium, rectangular cookies with hard icing designed to look like New York City metro cards. Although the MTA MTA has written letters to many of the trademark violators, it lacks the legal l egal staff  to go to court to enforce its trademarks and has allowed the value of those trademarks to deteriorate.58 Start-ups face a greater challenge than large, established companies in developing an effective strategy toward the management of trade and service marks.

Because small, new companies are often cash constrained, they face the dilemma of whether challenging—and winning—a trademark infringement lawsuit is worthwhile. The start-up might win the suit against a deep-pocketed competitor  but be dri driven ven out of busi busines nesss by the leg legal al effor effort. t. Take, for exa exampl mple, e, the case of  Haute Diggity Dog, the maker of dog toys. They created dog toys shaped like handbags, called “Chewy Vuiton.” Louis Vuitton, makers of the handbags that Haute Diggity Diggity Dog was parodying, parodying, sued them for degrading degrading the value of  Louis Vuitton’s trademark. While Haute Diggity Dog won the lawsuit, it lost a lot of distributors because because Louis Vuitton Vuitton sent cease-and-desist cease-and-desist letters to to the retailers during the lawsuit, causing the retailers to stop carrying Haute Diggity Dog’s products.59

Domain Names Domain names are the names used on Web sites to identify an organization providing a good or service. As Figure Figure 7 shows, domain names names have become an increasincreasingly popular form of intellectual property protection, as companies do more and more business over the Web. Domain names are registered by the Internet Corporation for Assigned Names and Numbers (ICANN) to the first party to seek registration for a name. 60 As with trade and service marks, it is useful to conduct a search before trying to register a domain name to make sure that you can obtain the name that you’d like to use. You can do this at the ICANN ICA NN Web site (www.icann.org). (www.icann.org). The protection of domain names is similar to the protection of trade and service marks. However, two important distinctions exist. First, because geographic regions are not meaningful in cyberspace, companies in different places are not permitted to use the same domain name, though they are permitted to use the same trade or service mark.61 Second, unlike with trade or service marks, common words can be used

as domain names. For example, Procter & Gamble has obtained the domain name 62

“cavities.com.” Your domain name also cannot adversely affect another company’s business. If it does, the company whose business has been hurt can sue you for control of your domain name. For example, Universal Tube and Rollerform Equipment Corporation

224

 

Trade Secrets, Trademarks, and Copyrights

FIGURE 7 60,000,000

Use of Domain Names

54,048,396

55,000,000 50,000,000

46,412,165

39,363,493

45,000,000 40,000,000 35,000,000

32,259,223

30,000,000 27,701,020

28,801,916

25,000,000 20,000,000 15,000,000

10,008,475

21,000

30,000

71,000 240,000 828,000

4,038,744 2,292,000

10,000,000 5,000,000 0

199 993 3 19 1994 94 19 1995 95 19 199 96 19 1997 97 199 998 8 199 999 9 20 2000 00 20 200 01 20 2002 02 20 2003 03 200 004 4 20 2005 05 20 2006 06

This figure shows the dramatic rise in the number of domain names in use over time. con tained in http://www. /www.zooknic.com/Domains/dn_length.html. zooknic.com/Domains/dn_length.html. Source: Adapted from data contained

has sued YouT YouTube ube for the rights to the www.youtube.com domain name because the volume of people going to Universal Univers al Tube’s Tube’s Web Web site www.utube.com www.utube.com when looking for www.youtube.com www.youtube.com has caused Universal Tube’s Web Web servers to crash repeatedly repeatedly.. The enforcement of a domain name occurs in a similar way to the enforcement of a trade or service mark. If you believe that someone else has infringed on your domain name or has taken action to lower its value, you can sue the offending party. 63 For example, Sir Ratan N. Tata of Bombay, Bombay, India, the leader of India’s Tata group of companies (which includes Tata Steel, Tata Engineering, Tata Power, Tata Chemicals, Tata Finance, Tata Power, Power, Tata Tea, Tea, and Tata Sons Ltd) and the Sir Rata Tata Trust, Trust, sued a New Jersey porn site in 1999, and obtained an injunction against the latter’s use of  the Internet domain name Bodacioustatas Bodacioustatas.com .com because the New Jersey company’s company’s use of the domain name harmed the reputation reputation and, hence the value, value, of Sir Ratan’s companies.64 However,, enforcement of domain name infringement However infringeme nt is often more difficult than enforcement of trade or service mark infringement because domain names operate in cyberspace. As a result, it is often difficult to determine the legal jurisdiction in which to sue an offender, and when that jurisdiction can be determined, it is often a place that does not strongly enforce intellectual property laws, making it hard for you to stop the offending action or collect damages. 65  Key Points

• Trade and service marks can be obtained on any nondescriptive, nongeneric

word, number, symbol, phrase, color, design, or even smell that distinguishes products and services o f one company of of another. • the Trade and service marks provide a negativefrom rightthose and must be enforced through legal action, which is often more difficult for start-ups to undertake than for established companies to conduct.

225

 

Trade Secrets, Trademarks, and Copyrights

• The same mark can be used by more than one company if that use will not cause confusion amongst consumers about the provider’s identity, and does not dilutecan theobtain value of another party’s mark.mark by using the mark or by • You a U.S. trade or service registering it with the USPTO; however, registration facilitates your ability to obtain similar rights in other countries and allows you to sue to enforce your mark. • Trade and service marks are lost through cancellation proceedings, abandonment, or if they take on generic meaning. • Domain names are names used on Web sites to identify the organization providing a good or service; they are protected through legal action.

I NTERNATIONAL I S S U E S P ROPERTY

IN

I NTELLECTUAL

There is no such thing as an international copyright, trademark, or patent. You need to obtain that piece of intellectual property protection in each of the countries in which you want to use it. As Figure 8 shows, this means that some countries, like the United States, receive a large number of patent filings, many of which are from entities domiciled in other countries. Moreover, this also means that the type of protection that you get in different countries on these three types of intellectual property depends on the laws of those countries and the willingness of their governments to enforce them. Just because you can obtain a patent, trademark, or copyright in one country doesn’t mean that you can obtain the same protection in another. For example, Anheuser-Busch, which has owned the trademark “Bud” in the United States since 1876, was only recently able to use that trademark in Hungary. There, the name “Bud” belonged to the Czech beer maker, Budejovicky Budvar, which had objected to Anheuser-Busch’s 66

efforts to use it.

Differences in Intellectual Property Regimes While a complete discussion of the differences in intellectual property regimes across countries is beyond the scope of this text, as a technology manager or entrepreneur, you need to be familiar with three of the most important differences: first-to-invent versus first-to-file rules, policies on the timing of disclosure, and the requirement to manufacture. You also need to understand the differences between the intellectual property systems in developing and developed countries.

First to Invent 

The United States is one of only two countries in the world that awards a patent to the first person to invent a technology. Most countries award a patent to the first person to apply for one, regardless of who invented the technology first. Because of  this rule, being the first party to file a patent application will not get you a patent in the United States. If you want a U.S. patent, you have to prove that you invented the technology technolog y before before anyone else.

226

 

Trade Secrets, Trademarks, and Copyrights

FIGURE 8

Top 20 Countries for Patent Applications Applicati ons

Japan USA Korea China European Patent Office Germany Canada Australia   s   e   c    i    f    f    O    t   n   e    t   a    P

Russian Federation United Kingdom Brazil India France Mexico Norway Hong Kong Singapore Argentina Italy Poland

0

50,000

100,000 150,000 200,000 250,000 300,000 350,000 400,000 450,000 Number of Filings Number of Resident Patent Applications filed Number of Non Resident Patent Applications filed

There are major differences across countries in both the total number of patent applications filed and the proportion filed  by residents. /www.wipo.int/ipstats/en/statistics/patents/patent_report_2006.html#P104 stats/en/statistics/patents/patent_report_2006.html#P104_9303. _9303. Source: WIPO Statistics Database, http://www.wipo.int/ip

The first-to-invent rule also has an important implication for obtaining worldwide patent protection: Failure to act quickly to apply for patents in other countries might keep you from obtaining patents outside of the United States. 67

Disclosure  In the United States, inventors who publish information about an invention up to

one year before applying for a patent remain eligible to obtain one. However, in many European countries, and in Japan, prior publication at any time will keep you from obtaining a patent.68 So if you are in a business like pharmaceuticals, in which researchers generally publish the results of their research in scholarly journals, you need to be very careful about the timing of publication when seeking patents outside the United States.

227

 

Trade Secrets, Trademarks, and Copyrights

Requirement to Manufacture  The United States imposes no requirement that an inventor actually produce a product or service uses theonly patented However , in many countries, you obtain patent that protection if you invention. are willingHowever, to manufacture your product in may that country. If you aren’t willing to do this, it doesn’t make sense for you to go through the time and expense to obtain a patent in that country. country. It will only be invalidated.69

Intellectual Property in Developing Countries  Many developing countries, like China and India, do not enforce their intellectual property laws very vigorously. In fact, research has shown a positive correlation  between the strength of intellectual property protection in countries and their thei r level 70 of per capita income. This correlation exists because companies in developing and developed countries generate value in different ways. In developed countries, companies generate much of their value from intellectual property—things like software code, drugs, automo bile designs, digital digital music files, and so on—whereas, on—whereas, in developing countries, countries, companies textiles, generatetoys, most of machinery. their value71from production of physical products—things like and As athe result, in developing countries, few domestic companies lose anything from weak intellectual property laws, and many benefit  because those laws laws facilitate imitation of foreign companies’ companies’ IP. IP. Moreover,, many developing country governments use weak intellectual property Moreover laws to facilitate imitation of expensive foreign products by low-cost local producers, thereby lowering the cost of those products to their citizens. Weak enforcement of  pharmaceuticall patents, in particular, pharmaceutica particular, allows local companies to reverse engineer and copy foreign drugs, reducing their cost by over 90 percent. 72 For example, the Brazilian government forced Abbott Laboratories to allow a Brazilian company to make a generic version of its patented AIDS drug in return for a 3 percent royalty. This action allowed the Brazilian government to save several hundred million dollars in medical costs by making an expensive drug available for $0.68 a pill. 73 As a technology entrepreneur or a manager, you need to adapt your technology strategy to these weak intellectual property laws. As Figure Figure 9 shows, you need to be

FIGURE 9

Conflict over Patents

Many people in developing countries consider it unfair that biochemical companies from industrialized countries exert patent rights to genetically engineered seeds, which raises the price of the seeds used by their farmers. /www.thimmakka.org/N ka.org/Newsle ewsletters tters/neemi /neemii.htm i.html. l. Source: http://www.thimmak

228

 

Trade Secrets, Trademarks, and Copyrights

prepared for conflict with governments and businesses in developing countries over enforcing your intellectual property rights. Moreover, you need to expect thatAmerican your products will be in developing countries, where counterfeiting costs companies anpirated estimated $20 billion 74 to $24 billion per year. This is particularly true if your company produces an intellectual property–intensive product that is also easy to duplicate, like a piece of computer software or a music video. You might want to minimize your risk of loss by keeping valuable IP out of  developing countries. For instance, you might avoid conducting R&D in countries with weak records of protecting intellectual property. 75 However, you might want to take advantage of weak intellectual property laws in developing countries by designing a strategy in which the transfer of your technology to local companies helps your company to compete better. For example, Toyota is building its Prius hybrid car in China even though the Chinese government does not enforce patents vigorously because it is counting on imitation by Chinese manufacturers to drive down the cost of batteries, which will attract buyers to its hybrids. 76

International Agreements on Intellectual Property Several international agreements on copyrights, trademarks, and patents exist, and facilitate the international protection p rotection of intellectual property. The major international agreement on copyrights is the Berne Convention, which sets a minimum level of  copyright protection that has to be provided in signatory countries, and requires the same level of protection to be provided both to citizens and noncitizens. 77 (However, it does not set the actual level of copyright protection in different countries, nor does it provide for protection in multiple nations.) The major international agreement on trademarks is the Madrid Protocol, which gives any individual who is a resident or citizen of one of the 70 signatories to the agreement the ability to file for international registration in as many of the signatory countries as they want through their home country’s patent and trademark office. 78 Because the other country applications are given the same filing date as the home country application, this protocol has made the process of obtaining trademark protection around the world easier and less expensive. 79 Several international agreements cover patents. The European Patent Convention allows an inventor to apply for patents in all participating European countries through a single application. If the patent is granted, it is then enforceable in all the participating countries in accordance with each country’s laws.80 The Paris Convention, which was signed by over 160 countries, prohibits the provision of differential patent rights to citizens of a country, establishes priority in patent applications, and allows you to disclose your invention to file for another country’s patents. Consequently, in any signatory country to the Paris Convention, you will obtain exactly the same patent protection as any citizen of that country, don’t have to rush around the world to file your patents, and don’t have to time your

applications around the world to avoid disclosure problems that would preclude you from getting a patent.81 The Patent Cooperation Treaty allows inventors in the 100 signatory countries to apply for patent protection in one country and preserve the right to apply for patent protection in all other signatory countries for 30 months. Inventors can reduce their patent costs by waiting to see what happens to their patent applications

229

 

Trade Secrets, Trademarks, and Copyrights

at home before applying for patents in other countries because the likelihood that foreign patent applications will be granted is linked to what happens to the applica82

tion The in the home country. Trade-Related Aspects of Intellectual Property Rights Agreement (TRIPS) makes uniform, at 20 years, the length of patents in all signatory countries, requires the countries to provide patents on chemical and pharmaceutical products, 83 and restricts mandatory licensing.84 This agreement strengthens intellectual property laws in developing country signatories. For example, India used to recognize patents only on the process of making drugs, allowing companies to copy patented drugs by coming up with alternative ways to synthesize them, but to be part of TRIPS, India had to agree to enforce patents on pharmaceutical products.  Key Points

• There is no such thing as international copyrights, trademarks, or patents; all three of these forms of intellectual property protection must be obtained in each country where a company would like to have them. • Major differences exist across countries in laws governing patenting, including whether they award patents to the first to invent or the first to file, their policies on the timing of disclosure, and their requirement to manufacture. • Developing country governments often do not enforce intellectual property laws vigorously because companies in those countries generate little value from intellectual property and because weak intellectual property laws reduce the cost of many products. • Weak intellectual property laws in developing countries lead to widespread piracy,, and require companies to formulate strategies that are effective under piracy such conditions. • Several international agreements make it easier to obtain intellectual property protection in multiple countries; the most important of these agreements are: the Berne Convention, the Madrid Convention, the European Patent Convention, the Paris Convention, the Patent Cooperation Coo peration Treaty, Treaty, and the Trade-Related Aspects of Intellectual Property Rights Agreement.

DISCUSSION QUESTIONS 1. What types of legal tools can be used to protect intellectual property? How should you choose among them? When is each of them effective? 2. Why does secrecy work better better for some products and processes than for others? Why does it work  better in some industries than than in others? 3. What are the advantages advantages and disadvantages of  using copyrights to protect software? Why do these

an infringement of those rights? Explain your position. 6. What, if anything, should be done to protect protect the rights of owners of copyrighted video and audio recordings, given the development of file-sharing technologies? Explain your reasoning. 7. China has relatively weak intellectual property laws, allowing many Chinese companies to imi-

advantages and disadvantages exist? 4. When can a company company hire an employee who worked at another company to develop a new product? What factors influence when this can be done? 5. Do you think that what peer-to-peer peer-to-peer digital filesharing networks do is “fair use” of copyrights or

tate products and services developed in other countries. To enter the World Trade Organization (WTO), China needs to strengthen its intellectual property laws. Is it better for China to strengthen its intellectual property laws and enter the WTO or not strengthen its laws and remain outside the WTO? What factors influence your evaluation?

230

 

Trade Secrets, Trademarks, and Copyrights

KEY TERMS Berne Convention: An international agreement that sets a minimum level of copyright protection in signatory countries and requires the same level of protection to be provided both to citizens and noncitizens. Consideration: The provision of something of value in return for agreeing to take a costly action or to make a purchase. Copyright: A form of intellectual property property protection provided to the authors of original works of authorship. Domain Name: The name used on Web sites to identify an organization providing a good or service. European Patent Convention: An agreement that allows an inventor to apply for patents in all participating European countries through a single application. Madrid Protocol: The major international agreement on trademarks; it gives any individual who is a resi-

Observable-in-Use: A condition under under which the intellectual property underlying a technology can be observed when it is used. Paris Convention: An intellectual property agreement signed by 160 countries that prohibits differential patent rights to citizens and establishes priority in patent applications. Patent Cooperation Treaty: An agreement by 100 countries to preserve the right to apply for patent protection in signatory countries for 30 months. Tacit Knowledge: Knowledge about how to do something that is not documented in written form. Trademark: A word, number, number, symbol, phrase, color, color, design, or even smell that distinguishes the products and services of one company from those of another. Trade-Related Aspects of Intellectual Property Rights

dent or citizen of one of the 70 signatories the ability to file for international registration in as many of the signatory countries as they want. Noncompete Agreement: A legal document that bars a person from working for a competitor for a period of time after his or her employment has ended. Nondisclosure Agreement: A legal document in which a person agrees not to make private information public.

Agreement (TRIPS): An agreement that makes uniform, at 20 years, the length of patents in all signatory countries, requires the countries to provide patents on chemical and pharmaceutical products, and restricts mandatory licensing. Work for Hire: Work that is done under the scope of a person’s employment or under a written agreement requesting that the work be done for someone else.

PUTTING IDE DEAS AS

INT IN TO

PRACTICE

1. Deciding on a Trade Secret Pick a company that you are interested in investigating. Identify a new product or process that the company is developing

Step 3: Once you have identified something that you can trademark, outline the steps that you’ll take to obtain a trademark on it.

or might develop. Using the information presented in this chapter, evaluate whether the company should use trade secrecy to protect the technology. Make sure to explain the pros and cons of that choice, and why you make this recommendation. 2. Obtaining a Trademark The purpose of this exercise is to identify a trademark for a business. Pick a  business that you know well. Then follow the the steps  below to identify a trademark. trademark.

Step 4: Explain how the trademark will help your business. What will the trademark protect? How will you enhance the value of the trademark? 3. Obtaining a Domain Name The purpose of this exercise is to learn how to obtain an Internet domain name for a business. Pick a business that you know well. Then follow the steps below to obtain a domain name for it.

Step 1: Identify a word, number, number, phrase, symbol, design, or combination thereof that you would like to trademark. Step 2: Go to www.uspto.gov www.uspto.gov and search existing

Step 1: Identify the word or words that you would like to use as your domain name. Step 2: Go to www.icann.org www.icann.org and search existing

trademarks to make sure that the word, number, phrase, symbol, or design you selected is not already trademarked. If so, evaluate whether you would be allowed to use the same word, phrase, number,, symbol, or design. If your evaluation is number positive, explain why. why. If your evaluation is negative, select something else to trademark.

domain names to make sure that the one you selected is not already being used by someone else. If it is, select a different domain name. Step 3: Once you have identified a domain name that you can obtain, please outline the steps that you’ll take to obtain the domain name for your  business.

231

 

Trade Secrets, Trademarks, and Copyrights

NOTES 1. Adapted from Cringely, Cringely, R. 1992. Accidental Empires. New York: Harper Collins. 2. Gleick, J. 2004. Get out of my namespace. New York  Times Magazine, March 21: 44–49. 3. Afuah Afuah,, A. 2003. Innovation Management. New York: Oxford University Press. 4. Nelson, R., and S. Winter Winter.. 1982. An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap Press. 5. Teece, D. 1998. Capturing value from knowledge assets: The new economy, economy, markets for know-how and intangible assets. California Management Review, 40(3): 55–79. 6. Zucker Zucker,, L., M. Darby, and M. Brewer. 1998. Intellectual human capital and the birth of U.S.  biotechnology enterprises. enterprises. American Economic

24. Guth, Microsoft sues to keep aide from from Google. 25. Woolf, Non-competition agreements. 26. Hart, M. 1996. Palm Computing Computing Inc. (A), Harvard Business School Case, Number 9-396-245. 27. Silverman, A. 1997. Understanding copyrights: Ownership, infringement, and fair use. JOM, 49(8): 60. 28. U.S. Department of of Commerce. 1992. General Information Concerning Copyrights. Washington, DC: U.S. Government Printing Office. 29. Austen, I. 2005. Block by block toy competitors competitors  build a case against Lego. The Plain Dealer, May 20: G1, G5. 30. Mesa, P., P., and R. Burgelman. 2004. Finding the balance: Intellectual property in the digital age. In G. Burgelman, C. Christiansen, and S. Wheelwright

Review, 88(1): 290–305. 7. Winter Winter,, S. 2000. Appropriating the gains from innovation. In G. Day and P. P. Schoemaker (eds.), Wharton on Managing Emerging Technologies Technologies . New York: John Wiley. 8. Teece, Capturing value from knowledge assets. 9. Chally Chally,, J. 2004. The law of trade secrets: Toward Toward a more efficient approach. Vanderb Vanderbilt ilt Law Review, 57(4): 1269–1311. 10. Yoffie, D. 2003. Intellectual property and a nd strategy, strategy,  Harvard Business School Note, Number 9-704-493. 11. Chally Chally,, The law of trade secrets. 12. Schil Schilling, ling, N. 2005. Strategic Management of  Technological Innovation . New York: McGraw-Hill. 13. Chally Chally,, The law of trade secrets. 14. Allen, K. K. 2003. Bringing New Technology Technology to Market.

(eds.), Strategic Management of Technology and Innovation. New York: McGraw-Hill/Irwin. 31. U.S. Department of Commerce, Commerce, General Information Concerning Copyrights. 32. Silverman, Understanding Understanding copyrights. copyrights. 33. U.S. Department of Commerce, Commerce, General Information Concerning Copyrights. 34. McBride, S. 2006. Music industry industry sues XM over replay device. Wall Street Journal, May 17: B1, B10. 35. Kiron, D. 2001. Napster Napster.. Harvard Business School Case, Number 9-801-219. 36. Moon, Y. Y. 2005. Online music distribution in a postNapster world. Harvard Business School Case, Number 9-502-093. 37. Silverman, Understanding Understanding copyrights. copyrights. 38. Yoffie, Intellectual property and strategy.

Upper Saddle River, NJ: Prentice Hall. 15. Fitzpatrick, W., W., S. DiLullo, and D. Burke. 2004. Trade secret piracy and protection: Corporate espionage, corporate security and the law la w, Advances in Competitiveness Research, 12(1): 57–68. 16. Schreiner Schreiner,, B. 2005. Colonel’s recipe remains a secret. Columbus Dispatch, July 24: G1, 2. 17. Wingfield, N. 2006. At Apple, secrecy complicates life but maintains buzz. Wall Street Journal, June 28: A1, A11. 18. Allen, Bringing New Technology Technology to Market. 19. http://www.crainsclevel /www.crainscleveland.com/ar and.com/article.c ticle.cms? ms? articleId=38171 20. Karush, S. 2005. IBM settles settles suit over intellectual

39. Moon, Online music music distribution in a post-Napster post-Napster world. 40. Associated Press. 2007. Sony Sony BMG to reimburse reimburse customers for CD damage. Wall Street Journal,  January 31: B4. 41. Smith, B., and S. Mann. 2004. Innovation Innovation and intellectual property protection in the software industry: An emerging role for patents. University of Chicago Law Review, 71: 241–264. 42. Barnett, W., W., and M. Leslie. 2006. Facebook. Stanford Graduate School of Business Case, Number E-220. 43. Yoffie, Intellectual property and strategy. 44. Silverman, Understanding Understanding copyrights. copyrights. 45. Smith and Mann, Innovation and intellectual intellectual prop-

property. The Plain Dealer, March 23: B2. 21. Guth, R. 2005. Microsoft Microsoft sues to keep aide from from

erty protection in the software industry. 46. Silverman, A. 2005. How to customize customize and maxi-

Google. Wall Street Journal, July 20: B1, B3. 22. Farkas, K. 2004. Patio enclosures enclosures awarded $8.6 million in suit. The Plain Dealer, August 31: D1. 23. Woolf, L. 2004. Non-competition agreements. FDCC Quarterly, Summer: 333–342.

mize federal trademark protection. JOM, 57(10): 72. 47. Clark, D. 2007. At the heart of Cisco’s iPhone lawsuit: A desire for open standards. standards. Wall Street Journal,  January 12: A9. 48. Gleick, Get out of of my namespace.

232

 

Trade Secrets, Trademarks, and Copyrights

49. Ibi Ibid. d. 50. U.S. Department of Commerce, Commerce, General Information Concerning Trademarks. 51. Silverman, How to customize and maximize federal federal trademark protection. 52. Schil Schilling, ling, Strategic Management of Technological Innovation . 53. Ethe Etherton, rton, S. 2002. Let’s Talk Patents. Tempe, AZ: Rocket Science Press. 54. Clark, At the heart heart of Cisco’s iPhone lawsuit. lawsuit. 55. Cohen, D. 1991. Trademark Trademark strategy revisited.  Journal of Marketing, 55: 46–59. 56. Kopp, S., and T. Suter. Suter. 2000. Trademark strategies online: Implications for intellectual property protection. Journal, 19(1): 119–131. 57. Allen, Bringing New Technology Technology to Market. 58. Chan, S. 2005. You You can take the A train, but don’t take its logo. You may get a warning letter. New York  Times, June 5: 27. 59. Flandez, R. 2006. Tiny firm wins “chewy Vuiton” Vuiton” suit, but feels bite. Wall Street Journal, November 28: B1, B5. 60. Vermette, N. 2000. Domain names in the realm of  trademark law. FICC Quarterly, 51(1): 1–15. 61. Ibi Ibid. d. 62. Kopp and Suter, Trademark Trademark strategies online. 63. Bagby Bagby,, J., and J. Ruhnka, 2004. Protecting domain domain name assets. The CPA Journal, 74(4): 64–67. 64. Gleick, Get out out of my namespace. 65. Bagby and Ruhnka, Protecting Protecting domain name assets. 66. Credeur Credeur,, M. 2006. Case of Bud finished. finished. International  Herald Tribune Tribune, May 25: 17.

67. Kesan, J. 2000. Intellectual property protection protection and agricultural biotechnology biotechnology.. American Behavioral Scientist, 44(3): 464–503. 68. Yoffie, Intellectual property and strategy. 69. Schil Schilling, ling, Strategic Management of Technological Innovation. 70. Varian, H. 2005. Copying and copyright. Journal of  Economic Perspectives, 19(2): 121–138. 71. Fishman, T. T. 2005. Manufacture. New York Times  Magazine, January 9: 40–44. 72. Bellman, E. 2005. India senses senses patent appeal. Wall Street Journal, April 20: A20. 73. Benson, T. T. 2005. Brazil to copy AIDS drug made by Abbott. New York Times, June 25: B12. 74. Fishman, Manufacture. 75. Anonymous. 2005. Drug Drug industry invests less in China, India. Wall Street Journal, November 10: D6. 76. Fishman, Manufacture. 77. Schil Schilling, ling, Strategic Management of Technological Innovation. 78. Ibi Ibid. d. 79. Haleen, I., and A. Scoville. Scoville. 2003. United States ratifies the Madrid Protocol: Pros and cons for trademark owners. Intellectual Property and Technology Law  Journal, 15(4): 1–3. 80. Kesan, Intellectual property property protection protection and agricultural biotechnology. 81. Schil Schilling, ling, Strategic Management of Technological Innovation. 82. Ibi Ibid. d. 83. Yoffie, Intellectual property and strategy. 84. Kesan, Intellectual property property protection protection and agricultural biotechnology.

233

 

234

 

Capturing Value from Innovation